Exhumation

In some quarters, David Hume’s attack on the theistic proofs is still considered definitive. Let us briefly consider two prongs of his offensive:
"A mind, whose acts and sentiments and ideas are not distinct and successive; one, that is wholly simple, and totally immutable; is a mind which has no thought, no reason, no will, no sentiment, no love, no hatred; or in a word, is not mind at all. It is an abuse of terms to give it that appellation, Dialogues Concerning Natural Religion" (Bobbs-Merrill, 1979), 159.

"Why not become a perfect anthropomorphite? Why not assert the Deity or Deities to be corporeal, and to have eyes, a nose, mouth, ears, &c.?...This world, for aught he knows...was only the first rude essay of some infant Deity...[or else] the production of old age and dotage in some superannuated Deity," Ibid., 169.

Now, what is striking about these two objections is that they appear to cancel each other out. For both objections seize on the issue of anthropomorphic theology. But they come at it from opposite ends of the scale. In the first stated objection, Hume attacks the classic concept of God because it is too theomorphic. Unless the divine mind subdivided into successive mental states, like our own, then it is no mind at all.

Yet, in his second objection, Hume reverses himself and attacks theology for being essentially anthropomorphic. According to him, the classic concept of God is just a scaled up version of man, except that it stops arbitrarily short of pure anthropomorphism.

Now, what are we to make of such contrary criteria? Let us go back and examine each of these objections separately.

I. Theology is too theomorphic

i) There are at least a couple of ways of construing Hume’s objection:
(a) The theologian begins with the human mind as his model, and then negates certain properties—while raising others to a higher power. But the act of negation renders the operative concept vacuous.
(b) Unless there is sufficient analogy between the human mind and its divine analogue, we have no basis of comparison. Perhaps (b) is a variation on (a).

By way of reply:

ii) Does the theologian begin with the mind of man, and then proceed to modify that model? Many theologians begin with the revelation of God in Scripture. Their concept of God is not abstracted from human psychology.

To this, Hume might counter in various ways:
(a) He is directing his attack against natural theology, not revealed theology.

Very well. But this omission leaves a large chink in his armor. After all, many Christians do turn to the Bible for their concept of God.
(b) Or Hume might allege that the Bible presents the same difficulty, only at more or more removes from the problem. In other words, he might contend that since the Bible is uninspired, the Bible writers were natural theologians, and so a theologian or ordinary believer who derives his concept of God from this second-hand source is still vulnerable to the same critique.

But for Hume to press that assault, he would need to engage the case for the inspiration of Scripture. Moreover, most Bible writers don't seem to be especially philosophical in their methods, so Hume would have to chart a pretty roundabout course to argue that they abstracted their concept of God by a conscious process of introspection and negation. Perhaps Hume would say that the process was unconscious rather than conscious. Okay, but it’s hard to see how Hume is privy to the unconscious process of another thinker, and—in any event—the claim is becoming so remote and attenuated by all these little caveats that it is increasingly difficult to see what is left of the original objection.

iii) But even if we leave the Bible out of account, is it true that the natural theologian takes the human mind as his point of reference? To begin with, not every theologian has the same theory of knowledge. There is a broad division between philosophers and theologians roughly in the rationalist tradition (e.g., Augustine, Anselm, Descartes), and philosophers or theologians essentially in the empiricist tradition (e.g., Aquinas, Scotus, Reid).

Now, even if Hume’s criticism held against the rationalist version of natural theology, it is unclear how that carries over to the empiricist version—especially seeing as Hume himself subscribes to an empiricist epistemology.

For example, what if Aquinas were to say that he derives his natural concept of God, not from introspection, but observation—not from the mind, but the sensible world?

Perhaps, though, Hume would say that the empiricist version collapses into the rationalist version. For what Aquinas is doing in, say the teleological argument, is to draw certain inferences based on an assumed analogy between divine and human designs. The design inference is really a mental projection of a human designer, based on our experience with human means of manufacture.

Now, at one level, there is some truth to this characterization. A natural theologian does see a parallel between the handiwork of God and man-made artifacts. And so the natural theologian, whether a rationalist or empiricist, does take himself as the immediate standard of comparison.

But even if we grant all that, how does it constitute an objection to the theistic proofs? If the theistic proofs are anthropomorphic, so are the atheistic proofs. Hume is taking his own mind as the immediate point of reference, and saying that if the concept of the divine mind is incommensurable with his mental yardstick, then it isn’t a real mind at all.

It isn’t clear, then, how Hume’s criterion enjoys any advantage over the opposing position. For it seems to be the same criterion on both cases.

iv) Suppose that theistic arguments are anthropomorphic? What of it? The fact that the subject of knowledge takes himself as the proximate point of reference in knowing the object of knowledge is hardly limited to God as an object of knowledge. That applies to our understanding of the natural world and human society in general. I attribute a natural dam to a beaver. The beaver is the maker and designer of the dam. I likewise ascribe a beehive to a bee colony. I understand my fellow man by asking myself how I would think and feel and what I would do in the same situation. There is in all of this an element of analogy, of mental projection. But does that mean that a beaver is just a figment of my imagination, that sympathy is a subject without an object? All knowledge other than self-knowledge has this vicarious dimension.

v) Suppose that AI researchers succeeded. A smart computer would attribute its design to a computer designer. But suppose an atheistic computer got into a heated exchange with a theistic computer. The atheistic computer would complain that the design inference was computermorphic. Well, in a way, the viewpoint of a smart computer would be computermorphic. And yet the computer was made by a computer designer.

vi) Moreover, the computer designer put something of himself into the computer. AI, if at all possible, would only be possible because the computer designer is the exemplar of AI. So there would be an analogia entis to justify an argument from analogy.

vii) Hume might object examples of beehives and beaver dams beg the question, for that is not a design inference or argument from analogy, but simple observation. By way of reply:
(a) This objection confounds the order of knowing with the order of being. However we detect design, whether by inference or observation, does not subtract from the objective fact of the matter.
(b) Hume’s objection to the teleological argument lay in the claim that it is illicit to infer divine design on analogy with human artifacts. Now even assuming, which I would deny, that every teleological argument is an argument from experience or analogy, there are natural artifacts as well as human artifacts. Human technology and human intelligence are not the only source of design. Beavers and bees and many other insects are impressive architects in their own right. An agent is needed to produce a natural artifact, and the artifact in question exhibits an algorithmic complexity which presupposes some sort of blueprint.

(In Darwin's Dangerous Idea [Simon & Schuster, 1995], Daniel Dennett offered an algorithmic defense of Darwinism. But even if this were a valid interpretation of the theory, it leaves entirely out of account the origin of the algorithm itself—as though that was self-explanatory! Dennett's argument also commits the composition fallacy.)

And if it takes an agent to build a beehive or a beaver dam, then that makes us wonder about other natural objects or events of comparable complexity, such as crystal formation or the genetic code. Unlike the beehive and beaver dam, these have no apparent agent, but if it takes an agent to make a beehive, with its self-symmetrical design, then what about a polycrystal? If the one demands an agent to effect that design, why not the other?

Or is it a case of remote agency? An automated factory manufactures artifacts, but someone built the factory. Even in the case of a beehive, where the proximate agent is apparent, the bees are executing a design not of their own devising. The whole process is front-loaded with abstract information. Indeed, it is something of a puzzle even to know how the design specs for such an intricate operation can be packed away in the brain of a bee. It reminds me more of remote control signaling, like a kid with a toy airplane.

II. Theology is too anthropomorphic

Hume says that a timeless mind is not a mind at all. His statement takes the form of an assertion rather than an argument. It isn’t clear from Hume’s denial just what he denies: is his denial ontological or epistemic? Is he saying that a timeless mind is impossible? Or his he saying that such a mind would have so little in common with human experience that we don’t know what the proposition amounts to?
i) Certainly there are differences between a timeless mind and a finite mind. But the differences seem to be easily stated. In a timeless mind there would be no thought process, no discrete and successive mental states. Rather, it would be like one very long and indivisible thought. Take an infinite number series or set. It is all there, as a given totality.
ii) Why should reason be equated with a particular mode or process of reasoning? Even at a human level, it might take us a while to arrive at a certain state of knowledge, but once we know it, we know it. Is the process of coming to know something the essence of knowledge, or is it the result, of what is known? What is incoherent or unintelligible about the idea that God simply knows all things without discursive means?
iii) The difference between a timeless mind and a finite mind is not that a timeless mind is bereft of reason and will, but rather, that it takes no time for a timeless mind to know and to will. There is no interval between idea, intent, and execution.
iv) The real difficulty lies not so much with a timeless mind as it does with a finite mind. How can a finite mind discover an infinite number series or n-dimensional figure, like the Mandelbrot set? How does truth subsist, waiting to be discovered? How can only part of the truth be known or knowable unless someone knows the whole truth?

His objection to the argument from analogy follows along the same lines:

i) But this objection is so broad that it applies to any argument from analogy, and is not restricted to philosophical theology, per se. According to his strictures, I can’t compare a human dam to a beaver dam unless I attribute paddle-tail to the human engineer or a pair of pants to the animal engineer.

ii) Every argument from analogy is also an argument from disanalogy. What makes an argument from analogy a sound argument is not that it holds at every point of comparison, but that it holds at the relevant level of abstraction. Both a beaver and a man are creative agents which design artifacts.

iii) In objection, it may be said that there is a fatal equivocation in the comparison. When I compare a man to a beaver, I’m comparing the known to the known. Hence, there are direct external controls on the extent of the analogy. But the cosmological or teleological argument entails a comparison between the known and the unknown. Hence, there is no check on the procedure. The theistic proof smuggles in a question-begging assumption. By way of reply:

iv) This is not necessarily an argument from the known to the unknown. There is a difference between knowledge and proof. A theistic argument may be construed as giving reasons for tacit knowledge.

v) Even if we didn’t know the outer bounds of the comparison, the surplus works in favor of the theistic proof. If a house is an artifact, and if the universe is an artifact, then it would take vastly greater wisdom and power to design and effect a universe than it would to design and erect a house. Although that may fall short of Christian theism, it will burst the bounds of atheism.

vi) One office of reason is to reason from the known to the unknown. A good deal of science lies in extrapolating from the known to the unknown. And detective will attempt to infer the identity of the killer from the crime scene. Certainly there is a good deal more to the murderer than a fingerprint or drop of blood or strand of hair. Yet the disanalogy does not invalidate the value of detective work.

vii) The theologian can also say that the Bible presents a description of the divine attributes, and that, in turn, supplies a definition which delimits the range of the analogy.