Unnatural Acts

In a previous post about C.S. Lewis’s book Miracles we began to look at his treatment of the Natural vs. the Supernatural. In Chapter 3 Lewis rolls out the argument that serves as the necessary underpinning for the rest of the book; he calls it The Cardinal Difficulty of Naturalism.

It is important to remember that on Lewis’s view, nature is an interlocking, fully deterministic system. If at any point the current state of this unfolding process has not been completely determined by its previous state, then we have allowed the entry of something outside of nature. Therefore in a purely naturalistic view there is no room for free will – and of course free will is indeed a thorny problem. (Daniel Dennett has written two books on the subject – Elbow Room and Freedom Evolves – in which he ameliorates the problem by showing that the question as it is usually considered is itself barely coherent; they are very interesting books, but discussing them in any depth would be rather off topic here.) The one “loophole” in this view of nature that seems to be most attractive to consciousness theorists is quantum indeterminacy; both Arthur Young and the team of Roger Penrose and Stuart Hameroff have zeroed in on this as a way by which we escape the fetters of determinism. Lewis mentions this almost in passing (p. 21); he is clearly unconvinced that nature “plays dice”, as it were, and he does not see, in any case, quantum uncertainty as supernatural; he refers to it rather as subnatural, and then moves along. He does, however, make the rather arbitrary assertion that if we can have randomness sneaking into nature from below, then it is “quite on the cards” that there might be a door opening to the supernatural as well.

The problem that Lewis sees as an insurmountable challenge to naturalism is human reason. He argues as follows:

If naturalism is true, then all of our thoughts are material events, and are embedded, like everything else, in the interwoven fabric of nature. Because every natural event is caused by the preceding state of the natural world, that means our thoughts are links in a chain of Cause and Effect. But Reason, argues Lewis, is supposed to be something else altogether. If our thoughts are the result of Reason, then they supposed to be brought about by the logical relation of Ground and Consequent, not by the billiard-ball mechanics of Cause and Effect. He illustrates this with two different uses of the word because:

Grandfather is ill today because he ate lobster yesterday.

Grandfather must be ill today because he hasn’t got up yet.

In the first instance, because refers to the mechanical linkage between the consumption of lobster and Grandfather’s indisposition. In the second, his failure to get up is the cause of our belief that he is ill. In other words, Lewis says:

The one indicates a dynamic connection between events or “states of affairs”; the other, a logical relation between beliefs or assertions.

Now because, on the Naturalist view, our thoughts are events in Nature, then they must follow our previous brain-states as Effect follows Cause. But for a train of thoughts to have any meaning, they must follow one another as Consequent follows Ground. But, Lewis argues, the two systems are wholly distinct, and any correlation between them cannot be trusted.

This, then, is the point: A naturalist view simply cannot provide us with any basis for trusting our own reason, and any attempt to use such reason to defend the naturalist view has in effect sawn off the branch upon which it perches. Because of this, then, Lewis asserts that acts of Reason are gateways by which the Supernatural enters the natural world.

Lewis’s argument is, I think, in many ways quite sound. It is quite possible in principle that we cannot have ultimate philosophical certainty as to the reliability of our reasoning if the Naturalist view is correct. But Lewis counters the evolutionary argument, in which we might suggest that natural selection drives our cognitive apparatus in the direction of better reasoning, thereby converging on ever-more-reliable truth-detectors, by making a rather flat assertion:

Now natural selection could operate only by eliminating responses that were biologically hurtful and multiplying those which tended to survival. But it is not conceivable that any improvement of responses could ever turn them into acts of insight, or even remotely tend to do so.

Is that really true? Not conceivable? Perhaps not to Lewis, but it is abundantly conceivable to plenty of others, including me.

But as I said above, I will concede Lewis’s point: it is perfectly defensible to argue that the split between Cause and Effect, as must happen in our neural circuitry, and Ground and Consequent, as is required for logical rigor, is sufficient to undermine our certainty that our reasoning is ultimately sound. For Lewis, this is sufficient cause to dismiss naturalism! But another way to look at it is to say that while the abstract certainty that Lewis the philosopher hopes for might elude us, still we have ample pragmatic justification to continue to rely on our reason. To name just one example (one doesn’t need to look far), the description of the mechanical behavior of celestial bodies that we have used our reason to develop corresponds well enough, apparently, to the truth of the world that we can fling our space probes across the solar system with astonishing accuracy.

In other words, Lewis’s argument seems to be:

A) Naturalism implies complete determinism. Any non-determined events are instances of something outside of nature poking through.
B) If naturalism as defined above is true, then human Reason cannot be trusted with certainty.
C) If we cannot with complete certainty rely on our Reason, then none of our inferences, including any arguments in favor of naturalism, can be considered proven.
D) Therefore naturalism cannot be proven true.
E) Reason must derive from a Supernatural source.

I’d say the jury’s still out on all of this. Certainly the argument, as summarized above (and I hope I am doing it justice) is far from airtight. A) might be an overly narrow definition, and of course D) does not necessarily imply E). Lewis might, of course, be right anyway: human reason might indeed be supernatural, and might derive directly or indirectly from God, who might or might not exist. But the derivation of human reason from God is itself an enormous and otherwise unsupported assertion. It seems to me that Lewis’s blind spot is that, with his philosopher’s training, he simply cannot abide the naturalistic implication that there are possible weaknesses in the foundation of Reason itself; for him this constitutes a sort of proof that naturalism must be wrong. What he does not seem willing to accept is that Naturalism might be true, but might also be, by its very nature, unprovably so, and that our Reason can be good enough without being perfect.

Related content from Sphere


  1. Kevin Kim says

    I’ve seen the question put forth in some books I’ve read: how does the existence of quantum indeterminacy help advocates of free will? Actions arising from random causes are no more free than actions that arise deterministically.

    What’s the “indeterminist’s” response to the above question?


    Posted April 12, 2006 at 1:21 am | Permalink
  2. Malcolm says

    Hi Kevin, and thanks for visiting.

    If the events arising from quantum indeterminacy are always truly random, then right you are. Some suggest, though (for example, take a look at these Young and Penrose/Hameroff links), that if behavior at the quantum level can be constrained (Penrose and Hameroff suggest a sort of “orchestrated” waveform reduction) it would present a means by which human consciousness, or even God, might exert an influence on macroscopic events.

    There aren’t a lot of loopholes, but this one isn’t definitively closed yet.

    Posted April 12, 2006 at 11:24 am | Permalink