Nagel and his critics, Part X

It’s time at long last to bring my series of posts on the critics of Thomas Nagel’s Mind and Cosmos to a close, before it becomes a lot longer than the book itself.  There isn’t, in any event, much more to say about the naturalist critics, most of whom raise objections similar to those on which I’ve already commented.  But I’ve long intended to finish up the series with a post on reviewers coming at Nagel’s book from the other, theistic direction.  So let’s turn to what John Haldane, William Carroll, Alvin Plantinga, and J. P. Moreland have said about Mind and Cosmos.

Though objecting to materialist forms of naturalism, Nagel agrees with his naturalist critics in rejecting theism.  All of the reviewers I will comment on in this post think he does so too glibly.  Naturally, I agree with them.  However, as longtime readers of this blog know, the arguments and ideas often lumped together under the “theism” label are by no means all of a piece.  Thomists and other Scholastics develop their conception of God and arguments for his existence on metaphysical foundations derived from Aristotelian and Neoplatonic philosophy.  But most contemporary philosophers of religion do not, relying instead on metaphysical assumptions deriving from the modern empiricist and rationalist traditions which defined themselves in opposition to Aristotelianism and Scholasticism.  This is a difference that makes a difference in the reviews of Nagel now under consideration.  Haldane and Carroll, like me, are Thomists, and their approach to Nagel reflects that fact.  But the objections raised by Moreland and Plantinga are to a significant extent different from the sort a Thomist would make.

Moreland and Plantinga

Let’s consider the latter first.  Moreland reviewed Nagel’s book in the Winter 2012 issue of Philosophia Christi, Plantinga in the December 6, 2012 issue of The New Republic.  Here’s how they describe the debate between theism and atheism.  Plantinga characterizes theism as “the belief that there is such a person as the God of the Abrahamic religions” and atheism as the view that “there is no such person as God or any other supernatural being.”  He also tells us that God is “the premier person and the premier mind.”  What Nagel fails to see, Plantinga says, is that while materialism “cannot account for the appearance of life, or the variety we find in the living world, or consciousness, or cognition, or mind,“ theism by contrast “has no problem accounting for any of these.”

How so?  Plantinga writes:

As for life, God himself is living, and in one way or another has created the biological life to be found on Earth… As for the diversity of life: God has brought that about, whether through a guided process of evolution or in some other way.  As for consciousness, again theism has no problem: according to theism the fundamental and basic reality is God, who is conscious...

And so forth.  Moreland’s approach is more detailed (since he is writing for an academic journal) but very similar.  He speaks of making a “cumulative case for God” on the basis of the same “evidence” that Nagel tries to explain in non-theistic terms, and says that “the theistic argument can be understood as an inference to the best explanation.”  Continuing in this vein he writes:

[T]he alleged limits of appealing to theistic intentions are, in fact, what characterize an appeal to any unobservable, theoretical entity (for example, a quark) -- we attribute to that entity what and only what is needed to explain the data.  This alleged limitation is also characteristic of personal explanation.  We attribute to a person those and only those intentions needed to explain his behavior.  Moreover, in the case of God, we have other factors -- for example, religious experience, revelation, other arguments in natural theology -- that allow us to fill out God’s intentions for bringing our cosmos and us into existence.  (p. 432)

Now the way all this strikes me, and I think the way it would strike any atheist reader, is as follows.  There are, Plantinga and Moreland seem to be arguing, several phenomena which might at least in principle be given a non-theistic explanation but for which theism is in fact a better explanation.  And “theism” is understood as a hypothesis postulating an unobservable theoretical entity which instantiates such properties as the property of being a person, the property of being alive, the property of being conscious, etc.  But it instantiates them in a way that is different from the way that natural things do, making it a member of the class of “supernatural beings.”  But it is not just any old member of this class, but the “premier” member.  The way this is supposed to explain the evidence in question is as follows.  Since God, like us, instantiates properties like being alive and being conscious, he could plausibly be what imparted those properties to us if he had the intention of doing so.  That he did have such intentions is something Christian revelation and the like tells us.  And so on.

Now I am not certain that Plantinga and Moreland would accept this summary without qualification.  Perhaps there are aspects of it that they would rephrase.  And obviously they would add a great deal in the way of argumentation for theism so construed.  But as I say, this is what strikes me as a natural reading, and how I think it would strike the typical atheist reader.  And with all due respect to Plantinga and Moreland, I have to say that if this is what I thought theism and the “case” for theism really amounted to, I wouldn’t find theism any more philosophically interesting or challenging than Nagel and his critics do.  (Indeed, it pretty much is what I thought theism amounted to in my atheist days.) 

The main reason is not so much that I think Plantinga and Moreland fail to show that the existence of God, so understood, is sufficiently probable, though their remarks do smack of a “god of the gaps” approach.  The main reason is that God so understood is in my view not terribly philosophically interesting, and in particular not terribly God-like.  Certainly the approach just sketched has nothing to do with the way Augustine, Anselm, Maimonides, Avicenna, Aquinas, and classical theists more generally have historically approached these issues.   For the classical theist, God is not “a person” or “a being” -- not because he is impersonal or lacking in being, but because he is not “an” anything.  He is not an instance of any kind.  He is not in any genus.  He does not merely participate in or “have” mind, life, existence, or anything else, the way we do.  In that sense he doesn’t “have properties.”  Indeed, he has no parts of any sort, but is absolutely simple.  If he were not, then he would be just one more piece of furniture in the universe among all the others, requiring an explanation of his own -- an explanation of why he instantiates the properties he does.  Even if he instantiated them in every possible world, if he were a substance distinct from his properties, or had an essence or nature distinct from his existence, he would not have the absolute metaphysical ultimacy that, for classical theism, is definitive of God and that only what is absolutely simple can have.  For the classical theist, God is unparticipated being or subsistent being itself, unparticipated goodness or goodness itself, and whatever else we can attribute to him can be attributed only in an unparticipated sense rather than as the instantiation of a property. 

Furthermore, the main arguments for God’s existence in the classical theist tradition are not mere “arguments to the best explanation” or “cumulative case” arguments, and the reasons have to do precisely with what God is for classical theism.  The Aristotelian approach to arguing for God’s existence holds that whatever is a mixture of potentiality and actuality -- and thus less than fully actual, and needing actualization -- can in principle be explained only by reference to that which is, already as it were, pure actuality and thus need not and could not have been actualized by anything else.  The Neoplatonic approach holds that whatever is in any way composite -- and thus in need of some principle to account for the combination of its parts -- can in principle be explained only by reference to what is absolutely simple and thus need not and could not have any parts needing combination.  The Thomistic approach holds that anything whose essence is distinct from its existence -- and which thus could have failed to exist and requires something to impart existence to it -- can in principle be explained only by reference to that which just is existence itself, and thus need not and could not have had existence imparted to it.

God is, in short, precisely that in which everything else participates for its actuality or being.  To say that what is pure actuality or being itself is not the only possible explanation for the existence of other things but is still “the best explanation,” for which we might make a “cumulative case,” is a bit like saying of a certain triangle that its instantiating triangularity-as-such is not the only possible explanation for why it is a triangle, but is still the “best explanation,” for which we can make a “cumulative case.”  This gets the nature of both the explanans and the explanandum just fundamentally wrong.  The relationship between a particular triangle and triangularity-as-such is not a contingent one, and not something we know of via empirical hypothesis formation and the weighing of probabilities.  And neither is the relationship between the world and God -- between that which has participated existence and that which just is subsistent or unparticipated being -- of that sort.

Perhaps Plantinga and/or Moreland could agree with some of this, and not wish their remarks to be interpreted in a way incompatible with it.  On the other hand, Plantinga has famously been critical of the doctrine of divine simplicity, which is one reason Brian Davies classifies him as a “theistic personalist” who is at odds with classical theism.  (See Peter Weigel’s useful IEP article on divine simplicity for critical discussion of Plantinga’s objections.)   Both Plantinga and Moreland do in other contexts describe God as “exemplifying” properties.  What Moreland (writing with William Lane Craig) says about divine simplicity and immutability in Philosophical Foundations for a Christian Worldview is in my view also irreconcilable with classical theism.  And Plantinga’s and Moreland’s general metaphysical commitments are in any case very different from those of Thomists and other more classically inclined metaphysicians.

For instance, in arguing for theism Moreland makes heavy use of the notion of “possible worlds.”  While there is nothing wrong with possible worlds talk per se, its utility is, from a Thomistic point of view anyway, very limited.  In particular, it is no good to make claims about the nature or essence of a thing by trying to conceive what it might be like in various possible worlds, because in fact we can know what could conceivably be true of it in various possible worlds only by first knowing its nature or essence.  (I have addressed this issue before.  See chapter 1 of David Oderberg’s Real Essentialism for detailed discussion.)

Moreland claims:

Since our natures/capacities are contingent (they didn’t have to be this way), how is it that they are able to gain contact with the realm of necessary truths of, for example, logic and mathematics, when we can easily imagine worlds in which they fail to have this ability? (p. 435)

But is it in fact “easily imaginable” that rational animals (which is what we are) could lack the ability to know any necessary truths at all?  I would say it is not only not easily imaginable, but that it is impossible for a rational animal not to have the ability to know at least very basic necessary truths, such as the law of non-contradiction.  I would say that it is partially constitutive of rationality to have such an ability, so that to be a rational animal is ipso facto to have that ability.

Of course, it is easily imaginable that a rational animal could be either insufficiently mature or disabled in such a way that it cannot exercise the ability to know such necessary truths -- indeed, not only is it imaginable, it happens all the time (fetuses and people with severe brain damage being rational animals who cannot exercise this ability).  But that is very different from not having the ability.  It is just one example among others of the commonplace phenomenon of a thing not manifesting all the properties or proper accidents that flow from its essence (such as a dog, whose nature it is to have four legs, which is missing a leg due to an accident; an oak tree, whose nature it is to sink deep roots, which fails to do so because of a barrier that has been placed around it; and so forth). 

Moreland, it seems to me, has mischaracterized the problem he is addressing because he has been misled by what we think we can “imagine” in various “possible worlds” (say, a human-looking creature who never does even basic math -- which is simply not the same thing as a human being who lacks the ability in principle to do even basic math).  The problem isn’t: “Why do rational animals, who could in other possible worlds have failed to have the ability to know necessary truths, in fact have that ability in the actual world?”  The problem is: “Could rational animals, who could not have lacked the ability to know certain necessary truths, have arisen through purely material processes?”  The correct answer to that question, I would say, is: “No, not even in principle.”  (See e.g. my recent ACPQ article “Kripke, Ross, and the Immaterial Aspects of Thought.”)  And arguing for that answer has nothing to do with any appeal to possible worlds, or to what we can imagine or conceive, etc. 

Similarly, Moreland says that “the existence of immanent teleology” of the sort posited by Nagel is “surely an improbable state of affairs across relevant possible worlds” (p. 431).  As an Aristotelian, I would say that, on the contrary, the existence of immanent teleology is a necessary precondition of there being any natural substances at all, so that it is its non-existence that is not only improbable, but impossible, across relevant possible worlds.  For Thomists and other Aristotelians, immanent teleology is partially constitutive of what it is to be a natural substance in the first place.  For example, that an oak tree has an inherent tendency to sink roots into the ground -- which is an instance of immanent teleology or goal-directedness -- is just part of what it is to be an oak tree. 

Again, Moreland has in my view mischaracterized the problem he is addressing.  The problem isn’t: “Why do natural substances, which could have lacked immanent teleology in other possible worlds, in fact have it in this one?”  The problem is: “Could immanent teleology, which couldn’t possibly have failed to exist in the natural world, exist even in principle entirely apart from the direction of any intelligence whatsoever?”  Here too I would answer: “No, not even in principle.”  But the reason has nothing to do with possible worlds, “design filters” (to which Moreland also appeals), Paley’s watch, “specified complexity,” or any other such red herrings.  It has to do with the reasons summarized (though only summarized) in Aquinas’s Fifth Way, an argument I have developed and defended in several places (such as in Aquinas -- and which I expound and defend at much greater length than I have elsewhere in my article “Between Aristotle and William Paley: Aquinas’s Fifth Way,” forthcoming in Nova et Vetera).

It does take some philosophical work to show that immanent teleology presupposes a divine intellect, however, contrary to what Plantinga implies.  In his review of Nagel, he writes:

As for natural teleology: does it really make sense to suppose that the world in itself, without the presence of God, should be doing something we could sensibly call “aiming at” some states of affairs rather than others—that it has as a goal the actuality of some states of affairs as opposed to others? Here the problem isn’t just that this seems fantastic; it does not even make clear sense. A teleological explanation of a state of affairs will refer to some being that aims at this state of affairs and acts in such a way as to bring it about. But a world without God does not aim at states of affairs or anything else. How, then, can we think of this alleged natural teleology?

End quote.  I would say that this is simply too quick.  To see why, compare final causality and efficient causality.  We know from everyday experience that natural efficient causes have real causal power -- that heat really does melt ice, falling rocks really do break glass, and so forth.  Does this causal power necessarily derive from a purely actual uncaused cause, who imparts such causal power without itself deriving it from anything else?  Absolutely (so we Thomists would argue), for such causes are on analysis relevantly like the stick that moves the stone -- something which can bring about an effect (the motion of the stone) only insofar as it is being used to do so by something else (a hand, say, of which the stick is an instrument).  But it takes some philosophical work to see this -- it isn’t blindingly obvious just from the perception of melting, glass shattering, etc.  Similarly, we know that teleology is real just from our everyday awareness of natural regularities.  We know that heat “points to” melting things, that trees “tend toward” sinking roots, etc., just from everyday experience.  It doesn’t require a fancy “design inference,” a theory of “specified complexity,” or any other such hoo-hah to know this.  But showing that such natural teleology presupposes an ordering intelligence does take work.  And it takes it precisely because natural substances are not like watches or other artifacts, whose functions require a designer precisely because they are not built-in the way the natural tendencies of heat, oak trees, etc. are.  (I have, in an earlier post and in my First Things review of Plantinga’s book Where the Conflict Really Lies, criticized his tendency to conflate function -- which we do directly see in natural phenomena -- and design -- which we don’t, and which must therefore be inferred.  Though it is not to be inferred after the fashion of Paley and ID -- see my many posts critical of those approaches.)

Haldane and Carroll

This brings us to Haldane and Carroll.  Haldane reviewed Nagel’s book in the print edition of First Things, and Carroll in Public Discourse.  Haldane and Carroll commend to Nagel Aquinas’s approach to teleology, which is very different from the approach one sees in Paley, Plantinga, “Intelligent Design” theory, and the like.  Many suppose that you either have to see teleology as entirely extrinsic to the natural world -- the way the time-telling function of a watch is entirely extrinsic to it, existing only in the intentions of the watchmaker and the watch’s users -- or you have to see it as immanent to the natural world in a way that completely divorces it from any divine directing intelligence.  The former option is that of “Intelligent Design” theorists, who would effectively transform certain parts of natural science into the enterprise of trying to discern the intentions of a cosmic engineer.  The latter option is that of Nagel, who, understandably wary of the first alternative, maintains that theism need play no role whatsoever in accounting for the existence of teleology.

But this is a false choice.  To see why, consider again the analogy of efficient causes.  We can study what Scholastics call “second causes” without constantly adverting to the intentions of the “first cause.”  For instance, we can study chemical reactions and gravitation without constantly asking what God, as the source of all causal power, has in mind in making a world with chemical reactions and gravitation in it.  That is why atheists and theists can do chemistry and physics and get the same results.  To affirm God as “first” cause in the sense of being the source of all causal power is not say that only God ever really causes anything (as the occasionalist holds), and to say that second causes are true causes is not to say that they could operate even for an instant in the absence of God’s concurrence (as the deist holds).  There is, for Aquinas and other Scholastic philosophers, a middle-ground position: Because second causes are true causes, they have causal power that can be studied and partially understood all on their own, even if a complete understanding of their causality as such (as opposed to merely their causality qua specifically chemical phenomena, their causality qua specifically biological phenomena, etc.) requires reference to the first cause.

In the same way, we can we study the teleological processes immanent to natural phenomena without constantly adverting to a divine designer’s intentions.  The study of natural teleology is not a kind of disguised divine psychology.  To study the natural tendencies of tree roots is the study of a feature of trees themselves, not an indirect way of trying to read God’s mind.  Natural teleology really does depend upon the divine intellect directing things to their ends (as the Fifth Way shows, and contrary to Nagel’ position) but this does not mean that such teleology is not really built into natural things and knowable just by knowing the things themselves, whether or not one also affirms a designer (contrary to what ID supposes).  As Carroll writes:

An account of divine agency in the tradition of Thomas Aquinas would allay some of Nagel's concerns. Aquinas thinks that science begins not from God's intentions, but with an analysis of nature as we discover it. God creates a universe in which creatures, inanimate and animate, have their own particular natures, their own intrinsic principles of characteristic behavior. There is a real, albeit created, self-sufficiency to nature. This is so because God's continuing causal activity, which is necessary for the ongoing existence of whatever is, operates on a radically different level from the kind of causality that creatures exercise. In fact, even to speak of a different level of causality fails to capture the radical otherness of divine agency. It is true, as Nagel suggests, that theism understands the intelligibility of the world ultimately in terms of God's purpose, but this does not require, as he seems to fear, the denial of an intrinsic intelligibility in nature that the natural sciences discover.

Of course, whether this approach to teleology and to theism in general is correct cannot be settled without evaluation of the Thomistic arguments.  Haldane takes Nagel to task for not seriously engaging the arguments for theism -- Nagel being content for the most part to express his aversion to the idea of God.  As Haldane writes: “Saying ‘I don’t want to go there’ hardly counters the suggestion that this may be where the reasoning leads.”

In fairness to Nagel, though, the arguments of Aquinas and other classical theists, however much they have dominated Christian thought historically, are not the ones most likely to be heard by someone dipping into the contemporary philosophy of religion literature.  And you can’t follow even the best reasoning where it leads until you first hear it.  Here’s hoping that Nagel, who has at least in a rudimentary way recapitulated some of the themes of Aristotelian and Scholastic metaphysics, will at some point consider the best arguments in natural theology -- those of the Aristotelian and Scholastic traditions.