Reading Rosenberg, Part VI

Let’s continue our detailed critical look at Alex Rosenberg’s The Atheist’s Guide to Reality.  In the previous installment, we took a detour to consider how some of Rosenberg’s problematic views in the philosophy of biology are developed more systematically in his book Darwinian Reductionism.  Here we return to the text of Atheist’s Guide and to the subject of religion, though we are not quite done considering what Rosenberg has to say about biological matters.  For he argues that Darwinism not only makes theism unnecessary (as he falsely assumes), but is positively incompatible with it: “You can’t have your Darwinian cake and eat theism too,” insists Rosenberg.  In particular, he thinks Darwinism is incompatible with the idea that God is omniscient.  How so?

Obviously, everything depends on how one understands “Darwinism” and “theism.”  Rosenberg says dubious things about both.  He assures us that Darwinism put the final nail in the coffin of teleology, and that any theism worth bothering with must attribute to God the intention to create us, specifically.  He then reasons as follows: Since they are non-teleological, Darwinian processes do not aim at the generation of any particular kind of species, including the human species.  In fact the generation of any particular species, including us, is highly improbable.  No one could have known before the fact that evolution would go the way it did.  But then an omniscient God could not have used such processes as a means of creating us, since only a very foolish deity would think it likely that natural selection would result in intelligent life.  

One problem with this is that it is false to say that Darwinism is incompatible with teleology.  For one thing, it is by no means clear that Darwinism really drove teleology even out of biology, let alone the rest of the natural world.  To hear writers as diverse as Etienne Gilson, David Stove, James Lennox, J. Scott Turner, and Marjorie Grene tell it, Darwinism does nothing of the kind.  For another thing, even if we do suppose either that biological phenomena are entirely non-teleological, or that the teleological aspects of biological phenomena are real but can be reduced to non-teleological features, it doesn’t follow that there is no irreducible teleology in the natural world at all.  As I have noted many times, the Aristotelian-Scholastic tradition argues that irreducible teleology of a very basic sort -- namely, bare directedness toward an end -- must exist wherever even the most elementary kinds of efficient causation exists.  (Rosenberg’s problem, as I have noted before, is that like so many atheists he thinks of teleology entirely on the model of Paley’s watch -- where the teleology involves the functioning of parts relative to a complex whole, and where the function is imposed from outside on parts that would not otherwise have it -- instead of thinking of it in Aristotelian terms, in which teleology is intrinsic to natural phenomena rather than externally imposed, and where the functioning of parts in relation to a complex whole is only one, relatively rare instance of teleology among others.)

Now, I have myself argued that getting from the world to the God of classical theism requires the distinction between act and potency and thus (since the notion of potency goes hand in hand with the notion of finality) the existence at least at the bottom level of physical reality of immanent final causality or teleology.  So, to that extent it is correct to say that theism is incompatible with a non-teleological universe (or to be more precise, that the possibility of arguing from the world to the God of classical theism is incompatible with a non-teleological universe).  But Darwinism, on any construal, implies at most only the non-existence of certain kinds of teleology, not the non-existence of all teleology; indeed, on an Aristotelian view, Darwinian processes themselves, like all efficient causal processes, necessarily presuppose the reality of some deeper level of final causality.  But the existence of any actualization of potency at all, and of any immanent teleology at all, entail the existence of the God of classical theism, including all of the key divine attributes (such as omniscience).  That, at any rate, is what arguments like those summarized in the Five Ways purport to show when rightly understood and completely spelled out.  (I explain in Aquinas and in my ACPQ article “Existential Inertia and the Five Ways” how they have been spelled out and defended in the Thomistic tradition.  I suppose I need to point out to the uninitiated reader that the Five Ways are not, and are not intended to be, complete “stand alone” arguments, but only summaries of and starting points for lines of argument that are more fully developed elsewhere.)  Various objections might of course be raised against these arguments -- objections I have responded to at length -- but Darwinism per se isn’t relevant one way or the other.

To be sure, the arguments do not all by themselves show that the God whose existence they purport to prove intended us, specifically; by themselves they leave open the question of whether or not human beings are an accidental byproduct of natural evolutionary processes.  But this does not help Rosenberg, for five reasons.  First and most fundamentally, it is very odd for Rosenberg to claim that any theism worth bothering with must hold that God intentionally created the human race, specifically.  It’s true that some forms of theism (such as Christianity) hold that man was made in God’s image, but that claim is logically independent of the proposition that the God of classical theism exists.  (After all, those who claim that man was made in God’s image also hold that God existed before He created man, would have existed even if He had never created the human race, and would continue to exist even if He decided to destroy the human race.)  It is either intellectually sloppy or intellectually dishonest of Rosenberg to suggest that showing that the human race was not made in God’s image would suffice to refute theism per se.  If the God argued for in arguments like the Five Ways exists, then atheism is false and that is that.  The serious debate will be between forms of theism, not between theism and atheism.  Whether man is made in God’s image will be relevant to the question of which form of theism is correct, but not to the question of whether some form of theism or other is correct.

Second, those who say that human beings are made in God’s image do not mean that our bodily nature is made in God’s image.  What they mean is that our rational nature as thinking and willing creatures is a finite reflection of God’s nature; our bodily characteristics could have been radically different, consistent with our being made in God’s image in this sense.  Moreover, those who hold that we are made in God’s image also often claim, on the basis of independent philosophical arguments, that our rational and volitional powers cannot even in principle be accounted for in materialist terms, including Darwinian terms.  Many of them also hold that given its immaterial powers, the human soul must be specially created by God each time a new human being comes into existence.  (Contrary to the impression one might get from Rosenberg, this doctrine was not cooked up as a way to reconcile evolution with theism by concocting some aspect of human nature which Darwinian processes did not generate.  The doctrine of the special creation of the human soul is centuries old, and is a development of arguments for the soul’s immateriality that are in turn as old as Plato and Aristotle.)  So, those who regard man as made in God’s image do not have to maintain that a species genetically and/or phenotypically identical to homo sapiens sapiens was intended by God.  The most they need to maintain is that God intended some biological species or other to come into existence at some point or other to which rational souls might in principle be conjoined.  And it is at the very least much harder to maintain that Darwinism is incompatible with this claim, even on the most anti-teleological construal of Darwinism.  

Third, even if we supposed that God did have to intend a species with the particular genetic and phenotypic characteristics of homo sapiens sapiens, it does not in fact follow even from the most anti-teleological interpretation of Darwinism that this result was improbable -- or at least not improbable from the relevant, “God’s eye” point of view -- for reasons Rosenberg himself should have seen given his commitment to the multiverse hypothesis.  For suppose that, as multiverse proponents often suggest, the existence of our universe is not in fact as remarkable or improbable as it seems given that it is only one of an infinite number of parallel universes.  On this view, our universe’s existence seems remarkable and improbable to us only because of our vantage point within it; but in fact it was inevitable that it should arise out of a process in which every possible universe is generated.  In this case we can imagine that God, intending the existence of homo sapiens sapiens specifically, simply caused the multiverse to exist, knowing that, even though the evolution of our species in any particular universe would be improbable, it would be inevitable that it will arise in some universe or other.  (Not that I endorse this suggestion, mind you, since I don’t buy the multiverse hypothesis.  The point is that Rosenberg, who does buy it, is again being either sloppy or dishonest.)

Fourth, the probabilistic nature of Darwinian processes does not in any event exclude divine intervention within a particular universe, for reasons Elliot Sober calls attention to in his recent book Did Darwin Write the Origin Backwards?  That the probability of a tossed coin’s landing heads is 0.5 is, Sober notes, perfectly consistent with saying that its probability of landing heads is either 0 or 1.   For each probability is relative to background conditions.  Given only that the coin has been tossed, its probability of landing heads is 0.5; given that it has been tossed and that its upward velocity, air resistance, the surface on which it lands, etc. are of such-and-such a character, then (assuming determinism for the sake of argument), its probability of landing heads is either 0 or 1.  When we make the judgment that the probability of the coin landing heads is 0.5, we are ignoring hidden variables of the sort which, when factored in, would make the probability either 0 or 1. 

Now, when we say that mutations are random in the sense of occurring with equal probability whether or not they benefit an organism, this, Sober says, is like saying that a coin is equally likely to come up heads whether or not its doing so will benefit the gambler who is tossing the coin.  The latter claim is perfectly consistent with the fact that when all the hidden variables are taken account of, the probability of the coin coming up heads is either 0 or 1.  And the former claim is perfectly compatible with the fact that when all the hidden variables that determine a particular mutation are taken account of, its probability of occurring will be either 0 or 1.  But there is, Sober argues, nothing in Darwinian biology per se that entails that divine intervention cannot be among those hidden variables in certain cases.  (We might add that assuming otherwise is like assuming that the fact that human beings sometimes interfere with the course of random mutation and natural selection -- as we have in experiments on Drosophila -- shows that Darwinian processes never really occur in nature.)

But all of this concedes too much to Rosenberg in the first place, and that brings us to the fifth and final point, which is that Rosenberg’s entire argument rests on a crude misunderstanding of the nature of divine causality.  In particular, he evidently knows nothing about the traditional distinction between primary causality and secondary causality, and operates with a crudely anthropomorphic conception of deity.  For he assumes that making evolution compatible with theism would require supposing that God intervenes in biological history at various points in order to alter the course of events so as to ensure that homo sapiens sapiens comes about, but does so in so subtle a way that it looks like the product of random variation and natural selection (but really isn’t, which is why such an account of evolution would not be truly Darwinian).  

But this is like saying that the author of a novel has to “intervene” in the story at key points, keeping events from going the way they otherwise would in order to make sure that they turn out the way he needs them to for the story to work.  Indeed, it is like saying that the author of a science fiction novel in which such-and-such a species comes about via natural selection has to “intervene” at key points so as to make sure that the evolutionary process comes out the way he needs it to in order for the story to work -- but at the same time has to do so subtly so that none of the characters would guess that he had intervened in this way.  The very suggestion is silly, for the author isn’t one causal factor in the story among others.  His causality relative to the story is not at all like the causality of either the characters or the impersonal processes operating within the story.  

Similarly, on the classical theist conception of God, God is not one causal factor in the universe among others, not even an especially grand and powerful causal factor.  He is not a “first” cause in the sense of being followed in a temporal series by a second cause, a third cause, a fourth cause, etc.  Rather, He is “first” or primary in the sense of being the fundamental cause, the necessary precondition of there being any causality within the universe at all, just as the author of a story is the “first cause” of what happens in the story, not in the sense of generating effects in the way the characters and processes described in the story do, but rather in the sense of being the necessary precondition of there being any characters or processes in the story at all.  Things in the world are “secondary” causes, then, in the sense of deriving their being and causal power from God, just as the characters in the story have any reality and causality at all only because the author of the story has imparted it to them by virtue of writing the story.  (For more on primary and secondary causality, see this post and this post.)

Now, it would be absurd to suggest that Macbeth did not really murder Duncan, but that it was Shakespeare who committed the murder and merely made it look like Macbeth had done it.  This would be to treat the author as if he were a character in the story.  For the same reason, it would be absurd to suggest that in a science fiction novel in which such-and-such a species evolves, it is not really Darwinian processes that generate the species, but rather the author of the story who does so and merely made it seem as if Darwinian processes had done it.  But by the same token, it is absurd to suggest that if God creates a world in which human beings come about by natural selection, He would have to intervene in order to make the Darwinian processes come out the way He wants them to, in which case they would not be truly Darwinian.  This is to confuse primary with secondary causality, to think of God as if He were merely one causal factor in the world among others, like treating an author as if he were merely one character in his story among the others.  (Physicist Stephen Barr made this point well in his lecture at Franciscan University’s recent Science and Faith Conference.  As I have pointed out before, though, one shouldn’t push the author/story analogy too far.)

In short, Rosenberg thinks of God as a Paley-style watchmaker, an anthropomorphic tinkerer who cleverly intervenes in a natural order that could in principle have carried on without him -- in this case by manipulating evolutionary processes, like the Marvel Comics character the High Evolutionary.  But as I have so often emphasized when criticizing “Intelligent Design” theory, that is not the God of classical theism -- of Athanasius, Augustine, Anselm, and Aquinas, of Maimonides and Avicenna, or in general of Christian, Jewish and Islamic orthodoxy and of philosophical theism.  (I should emphasize for ID enthusiasts that none of this presupposes that the standard Darwinian story is in fact correct in all of its particulars, or even that it is correct at all for that matter.  For the point has nothing essentially to do with Darwinism or biology in the first place.  What Aristotelian-Thomistic critics of ID fundamentally object to is ID’s overly anthropomorphic conception of God and its implicit confusion of primary and secondary causality -- and that, by virtue of these features, ID muddies the waters in the debate between atheism and theism, fostering misunderstandings of the sort that Rosenberg and so many other atheists have fallen prey to.)