Reading Rosenberg, Part IV

Alex Rosenberg’s dubious use of physics was the focus of the previous installment of our look at his new book The Atheist’s Guide to Reality.  In this post we’ll look at his dubious biological claims.  “When physics disposed of purposes,” Rosenberg tells us, “it did so for biology as well.”  Now as I’ve noted before, in fact modern physics has not “disposed” of purposes at all, if what Rosenberg means by this is that physics has somehow established the metaphysical claim that the material world is devoid of objective teleological features.  All it has done is to make the purely methodological move of confining itself to non-teleological descriptions of the phenomena it studies.  This no more shows that teleology doesn’t exist than the fact that I am confining my comments in this post to Rosenberg’s work shows that no other philosophers exist.  Moreover, the non-teleological methodology of modern physics rules out irreducibly teleological explanations in biology only if you buy into Rosenberg’s “physics or bust” brand of scientism, which he has given us no good reason to do.

But Rosenberg does at least understand the implications of his position for the question of biological adaptation:

Scientism needs more than an explanation of this or that particular adaptation -- white fur in polar bears or the fact that bottom-dwelling fish have both eyes on the side of their bodies facing away from the bottom.  We need an explanation of how, starting from zero adaptations, any adaptation at all ever comes about.  The explanation we need can’t start with even a tiny amount of adaptation already present.  Furthermore, the explanation can’t help itself to anything but physics.  We can’t even leave room for “stupid design,” let alone “intelligent design,” to creep in.  If scientism needs a first slight adaptation, it surrenders to design.  It gives up the claim that the physical facts (none of which is an adaptation) fix all the other facts. (p. 50)

We might compare the alternative Rosenberg wisely rules out -- that the naturalist might allow in a tiny bit of adaptation at the start and build up more complex adaptations on that basis -- to the strategy of “homuncular decomposition” in the philosophy of mind.  This is the idea that we can explain human thought in terms of sub-personal homunculus-like processes which exhibit a lower degree of intelligence than the thought processes we are trying to explain; that these sub-personal processes can, in turn, be explained in terms of even “stupider” homunculi; those in terms of yet stupider homunculi still; and so on until we reach a lowest level of homunculi which are so stupid that their operations can be carried out by processes that are clearly purely material and non-intentional.  As critics like John Searle and John Haldane have pointed out, this whole procedure is fallacious.  The lowest-level homunculi will be carrying out operations that can intelligibly be said to “add up” to the higher-level mental ones only if they possess some minimum degree of intentionality, in which case the naturalist’s problem of explaining intentionality in non-intentional terms will merely be relocated rather than solved.  Proponents of the strategy fail to see this because, as Haldane notes, they confuse the intentional/non-intentional distinction with the more intentional content/less intentional content distinction.  And, I would add, if they were to bite the bullet and accept that there is genuine intentional content at least at some very low level of physical reality, they will have implicitly given up a physicalist conception of matter and revived an Aristotelian commitment to finality or “directedness” as a fundamental aspect of the natural order.

Rosenberg -- who correctly sees that to be a consistent naturalist in the philosophy of mind requires being an eliminative rather than reductive materialist vis-à-vis intentionality -- also realizes that to pursue a similarly reductive strategy in biology vis-à-vis teleology would be equally fallacious, and equally fatal to naturalism.  He sees that teleology and allied notions must be completely avoided, that smuggling in even a “slight” or “tiny” amount of adaptation would give the game away.  (As I discuss in The Last Superstition, Daniel Dennett is one naturalist who does not see this, or at least who constantly helps himself to teleological concepts which he cannot successfully “cash out” in naturalistic or non-teleological terms.  Dennett is also, as it happens, a well-known proponent of the homuncular decomposition strategy in the philosophy of mind -- in for a penny, in for a pound, and all that.)

Now, I would say that in fact you aren’t going to get any biological adaptation at all from a starting point utterly devoid of adaptation, any more than you are going to get the intentional from the non-intentional.  Biological adaptation is an inherently teleological concept, and the processes from which Rosenberg would derive it are (as he conceives of them, anyway) inherently non-teleological.  Reductive versions of materialism in the philosophy of mind are always disguised forms of eliminative materialism; they make use of mentalistic vocabulary while subtly but completely evacuating it of its ordinary mentalistic content.  And Rosenberg’s reduction of the adaptive to the non-adaptive does something similar.  But while Rosenberg’s eliminativism in the philosophy of mind is explicit, he does not make it clear that he is committed to an eliminativist position with respect to biological adaptation.  (Indeed, in his -- very interesting -- book Darwinian Reductionism, Rosenberg presents his position in philosophy of biology as reductionist rather than eliminativist.  I’ll have something to say about that book in a follow-up post.)

Here’s how the “reduction” goes in Atheist’s Guide.  Rosenberg sets the stage as follows:

Natural selection requires three processes: reproduction, variation, and inheritance.  It doesn’t really care how any of these three things get done, just so long as each one goes on long enough to get some adaptations.  Reproduction doesn’t have to be sexual or even asexual or even easily recognized by us to be reproduction.  Any kind of replication is enough.  (p. 59)

He later puts things instead by saying that in addition to “replication and variation… fitness differences [are] the last of the three requirements for evolution by natural selection.” (pp. 64-65).

With these criteria in hand, Rosenberg devotes several pages to sketching out scenarios in which inorganic molecules can be said to replicate, vary, differ in their fitness, and thereby give rise to “adaptation.”  And he has no trouble doing so given how broadly he construes the key concepts: The formation of crystals counts as an example of “replication”; the chemical difference between sugar and Splenda counts as an example of “variation”; an inorganic molecule’s being able to “persist or replicate or both” counts as “adaptation”; and so forth.

Thus does Rosenberg “show” how “adaptation” can arise from non-adaptation in a way that doesn’t “cheat” by smuggling in adaptation in at the beginning.  But this is a little like proudly proclaiming that you didn’t cheat on your exam, because the professor handed out the answers in advance.  It’s true, but only in a completely trivial and uninteresting sense.  For given how broadly Rosenberg is willing to allow us to construe the key notions, you might as well say that pebbles are “well-adapted” to their environment.  After all, they “replicate” (when one pebble is broken into two); they “vary” (the new pebbles are smaller than the original, and differ from it and from each other in shape); they “inherit” features from their parents (the new pebble is solid and rough, just like Dad -- a chip off the old block!); and they differ in their “fitness” (the new pebbles are smaller and thus less easily broken than their ancestors).  Descent with modification, in rock gardens no less than in botanical gardens!

But what does any of this have to do with organic phenomena, with biological adaptation?  Nothing at all; certainly Rosenberg does nothing to justify the claim that it does, other than to make the obligatory hand-waving reference to the Miller-Urey experiments and hydrothermal vents, and a passing concession to the effect that “molecular biologists don’t yet know all the details, or even many of them” about how organic processes might arise from inorganic ones.  Ah yes, all we’re missing is a few details.  Except that since the question at issue is whether biological adaptation can be explained in terms of the stuff about crystals, Splenda, etc., to leave out these “details” is just to fail to answer the question at all.  Rosenberg is like the guy who contracts to build you a house, clears the ground a little, and then takes off without doing anything else -- dismissing your concerns about the absence of a foundation, framework, walls, electrical, plumbing, etc. as mere quibbling over “details.”  Well, no, they’re not mere details.  They’re the house.  

What Rosenberg owes us is an account of how biological adaptation, specifically -- and not just the kind of “adaptation” a resilient inorganic molecule or a pebble exhibits -- can arise from physical processes that initially involve no biological adaptation at all.  And that means he owes us an account of what life is -- an account that makes it evident exactly how the sort of “adaptations” he describes add up to the kind a living thing exhibits.  But “What is life?” is a question which (oddly for a professional philosopher of biology) Rosenberg does not directly address.  He just speaks of “adaptation” sans phrase, and insinuates, without argument, that having given an account of processes that might in some extended sense of the word be called “adaptation,” he has thereby given an account of life.  

Now the Aristotelian tradition has, of course, an account of what life is.  Living things, it says, are those which exhibit immanent causation as well as transeunt (or “transient”) causation; non-living things exhibit transeunt causation alone.  Transeunt causal processes are those that terminate in something outside the cause.  Immanent causal processes are those which terminate within the cause and tend to its good or flourishing (even if they also have effects external to the cause).  For example, an animal’s digesting of a meal is a causal process that tends to the good or flourishing of the animal itself (though it also has byproducts external to the animal, such as the waste products it excretes).  By contrast, one rock’s knocking into another is a transeunt causal process, in that it does not in any sense tend to the good or flourishing of the rock itself.  (I had reason to address these matters in an earlier post.  I also address them in chapter 4 of Aquinas.  The best recent treatment of issues in the philosophy of biology from an Aristotelian point of view is in chapters 8 and 9 of David Oderberg’s Real Essentialism.  If you’re up for tracking down older Scholastic works on the subject, you might look for Henry Koren’s An Introduction to the Philosophy of Animate Nature or George Klubertanz’s Philosophy of Human Nature.)

Whatever the details, the Aristotelian conception of life is irreducibly teleological, for the notion of causal processes that tend toward the good or flourishing of the cause is itself inherently teleological.  (Though it is important to stress that the kind of teleology characteristic of living things is only one, relatively rare sort of teleology; and also that -- contrary to what Darwinian naturalists and “Intelligent Design” theorists alike suppose -- whether there is teleology in nature and whether there is a “designer” are separate questions.  On these issues, see my article “Teleology: A Shopper’s Guide” and my various posts criticizing ID theory from an Aristotelian point of view.  I should also note here that the “immanent causes vs. transeunt causes” distinction in Aristotelian philosophy of biology is different from the “immanent finality vs. extrinsic finality” distinction that arises in discussions of the question of whether final causality is immanent to the natural order -- as Aristotelians claim -- or whether it is entirely extrinsic or imposed from outside -- as moderns like Newton and Paley claim.)  

Rosenberg, who claims that teleology of any sort has been banished by physics and that “the physical facts fix all the facts,” will naturally have no truck with this conception of life.  But if the Aristotelian is correct to hold that life is an inherently teleological notion, then Rosenberg’s position vis-à-vis life is implicitly eliminativist: If life is inherently teleological but there is no teleology, then there are no living things either; there only seem to be.  That doesn’t mean, of course, that Rosenberg is committed to denying that there is a real difference between you and your corpse.  But it would mean that whatever important differences there are between you and a corpse, or you and the “adaptive” inorganic molecules Rosenberg describes, the difference does not involve your having some teleological or irreducibly “life-like” features that they lack.

But you needn’t be an Aristotelian to regard Rosenberg’s position on life as implicitly eliminativist.  Rosenberg implies that his account of adaptation is not merely an account of life that is consistent with scientism; he evidently regards it as the only account that is consistent with it.  Though he finesses the details, he is committed to the proposition that there cannot be any “adaptation” in the world that doesn’t boil down to the sort of thing he describes when he tells us his story about how Splenda is an example of “variation,” resilient inorganic molecules are instances of “fitness,” and so on.  Rosenberg’s Infallible Dogma -- “The physical facts fix all the facts!” -- cannot allow anything more than this; certainly it is hard to see how we could add any more to it without bringing in something like the Aristotelian notion of immanent causation and its attendant teleology.

Now, that means that biological adaptations must in Rosenberg’s view be entirely continuous with inorganic “adaptations” of the sort he describes.  And that means in turn that we are faced with a choice if we buy into Rosenberg’s premises.  We could say, on the one hand, that Rosenberg’s resilient inorganic molecules -- and, for all I know, pebbles too -- are “really” “living” things after all, even if very simple kinds of life.  (On this interpretation, when contemplating a crystal or a pebble, Rosenberg might soon find himself channeling Colin Clive, or maybe Gene Wilder.)  Alternatively, we could say that since inorganic molecules really are inorganic and so-called “living” things are not essentially different from them, it follows that so-called “living” things are “really” no more alive than inorganic molecules, pebbles, and the like are.  Again, this doesn’t mean that Rosenberg would have to deny that you differ significantly from a corpse or an inorganic molecule -- any more than, as an eliminativist about intentionality, he would deny that there is a significant difference between the words you utter and random noises.  But just as he would deny that the words you utter really have any meaning or semantic content that random noises lack, so too (on this latter interpretation) would he deny that there is any such thing as “life” which you have and the corpse or inorganic molecule lacks.

Given that Rosenberg clearly holds that the bottom-level physical facts determine what we should say about higher levels rather than the other way around, and given that he is already more than happy to eliminate other higher-level phenomena from our picture of the world, it is pretty clear that this latter, eliminativist view of life is the one Rosenberg is implicitly committed to.  And I don’t see why he couldn’t shrug his shoulders in agreement, given the other things he’s willing to say.  (Again, Rosenberg does in his book Darwinian Reductionism claim to be a reductionist rather than eliminativist vis-à-vis biology, but what he is directly addressing in that context is the question of whether functional descriptions pick out real features of the biological realm, rather than the question of whether there is a real distinction between living and non-living things.  Anyway, as I have said, I’ll discuss that book in a follow-up post.)

The problem with all of this isn’t that it’s absolutely bizarre, though of course it is.  The problem is that there is simply no reason whatsoever to take it seriously.  In particular, there is no reason at all to think that whatever is true of bacteria and bees, trees and toads, pigs and people is entirely exhausted by what physics tells us about these things, as Rosenberg understands “physics.”  It isn’t biological science, but ideological scientism that leads him to suppose otherwise.  And as we have seen, his sole argument for scientism is entirely devoid of merit.  Biological phenomena, as Rosenberg represents them, are like something you’d find in a taxidermist’s shop -- bits of dead matter stripped from a corpse and sewn onto a framework of cold wiring and stuffing, so as vaguely to look like something living.  Here as elsewhere, Rosenberg guides us, not to reality, but to a Frankenstein’s-monster-like simulacrum of reality.