One of the first philosophical arguments I ever encountered was C. S. Lewis' argument, found in his book Miracles, that naturalism is self-refuting because it is inconsistent with the validity of reasoning.1 The argument fascinated me, and as a young Christian I never missed a chance to present it in discussions with skeptics. I later discovered that this argument was the subject of the famous controversy with Elizabeth Anscombe, in response to which Lewis revised his argument.2 However, in my graduate studies in philosophy I discovered, with some exceptions, that the argument had received little attention even from Christian philosophers and was dismissed by many. As I analyzed the various ways in which the argument could be answered, the firmer the rock appeared on which it stood. And so when it came time to write my doctoral dissertation, I chose to defend this argument against naturalism. Even though my committee was solidly opposed to the conclusion of my argument, they nevertheless passed my dissertation.

I'm still persuaded that Lewis' argument is a good one. And the best way to honor Lewis' apologetic achievement, it seems to me, is not simply by repeating what he says, but by developing his ideas, asking probing questions of them, and developing the discussion in ways that reflects one's own thinking as well as Lewis'. The result of such refection on my part has persuaded me that, indeed, Christianity is credible for approximately the reasons that Lewis said it was. I believe that to a large extent not enough of this further reflection has been done in the secondary Lewis literature, and that is why I believe that despite Lewis' enormous popularity, subsequent apologists have underutilized the resources that he has provided for them.

All too often, Lewis' career as an apologist has been assessed biographically, focusing on Lewis himself rather than what he had to say. So, for example, it is often suggested that Lewis was profoundly upset by his exchange with Anscombe, and therefore he himself realized his apologetic arguments were inadequate. And this is frequently done without analyzing the points at issue in the debate. Thus a kind of "Anscombe legend" has shrouded Lewis' apologetic career, which in many minds seems to have outlived the actual arguments Lewis and Anscombe presented.3 The following passage, from Humphrey Carpenter's The Inklings, is a case in point:

Miracles was published in 1947. Early the following year, its third chapter, in which Lewis proved that human Reason is independent of the natural world, was publicly attacked at the Socratic Club, not by an atheist, but by a fellow Christian, the Catholic philosopher Elizabeth Anscombe. Lewis was unprepared for the severely critical analysis to which she submitted his arguments, for she proved in her turn that his "proof" was severely faulty. It is true that Lewis's most fervent supporters felt that she had not demonstrated her point successfully, but many who were at the meeting thought that a conclusive blow had been struck against one of his most fundamental arguments. Certainly after it was all over Lewis himself was in very low spirits. He and Hugo Dyson had organised an informal dining club with four of their pupils, Philip Stibbe, Tom Stock, Peter Bayley, and Derek Brewer, and the club happened to meet a couple of days after the Socratic duel. Brewer wrote in his diary: "None of us was at first very cheerful. Lewis was obviously deeply disturbed by his encounter last Monday with Miss Anscombe, who had disproved some of the central theory of his philosophy about Christianity. I felt quite painfully for him. Dyson said—very well—that now he had lost everything and had come to the foot of the Cross—spoken with great sympathy." Brewer adds that Lewis's imagery was "all of the fog of war, the retreat of infantry thrown back under heavy attack."

Lewis had learnt his lesson: for after this he wrote no further books of Christian apologetics for ten years, apart from a collection of sermons; and when he did publish another apologetic work, Reflections on the Psalms, it was notably quieter in tone and did not attempt any further intellectual proofs of theism or Christianity. Though he continued to believe in the importance of Reason in relation to his Christian faith, he had perhaps realized the truth of Charles Williams' maxim, "No one can possibly do more than decide what to believe."4

Now it is important to see what is missing in this discussion: any treatment of what the arguments of Lewis and Anscombe actually were. Apparently Carpenter thinks he can draw the conclusion that Anscombe had shown Lewis' argument to be "severely faulty" without analyzing the arguments themselves, simply based on biographical considerations. Nor is it indicated here that Lewis revised the argument in the 1960 Fontana edition of Miracles to meet Anscombe's objections. In A. N. Wilson's biography, the Anscombe legend is pushed to extreme lengths: the incident's psychological impact explains Lewis' "retreat" into children's fantasy (Narnia), the fact that a female witch offers skeptical arguments in The Silver Chair that nearly beguile the protagonists of that story, and the fact that Lewis enjoyed corresponding with American women (including Joy Davidman).5

John Beversluis, in his book C. S. Lewis and the Search for Rational Religion and in an essay for Christian History entitled "Beyond the Double-Bolted Door," also makes a good deal of the psychological impact of the Anscombe incident, and so uses the Anscombe legend to support his claim that Lewis' apologetics are woefully inadequate from a philosophical perspective.6 But he does analyze the actual arguments to support the claim that "the arguments that Anscombe presented can be pressed further, and Lewis' revised argument does nothing to meet them." And very much to his credit, in his subsequent review of Wilson's book, Beversluis abandons the Anscombe legend entirely:

First, the Anscombe debate was by no means Lewis's first exposure to a professional philosopher: he lived among them all his adult life, read the Greats, and even taught philosophy. Second, it is simply untrue that the post-Anscombe Lewis abandoned Christian apologetics. In 1960 he published a second edition of Miracles in which he revised the third chapter and thereby replied to Anscombe. Third, most printed discussions of the debate, mine included, fail to mention that Anscombe herself complimented Lewis's revised argument on the grounds that it is deeper and far more serious than the original version. Finally, the myth that Lewis abandoned Christian apologetics overlooks several post-Anscombe articles, among them "Is Theism Important?" (1952)—a discussion of Christianity and theism which touches on philosophical proofs for God's existence—and "On Obstinacy of Belief"—in which Lewis defends the rationality of belief in God in the face of apparently contrary evidence (the issue in philosophical theology during the late 1950s and early 60s). It is rhetorically effective to announce that the post-Anscombe Lewis wrote no further books on Christian apologetics, but it is pure fiction. Even if it were true, what would this Argument from Abandoned Subjects prove? He wrote no further books on Paradise Lost or courtly love either.7

Anscombe herself did not recall any devastating encounter, attributing the adverse reaction of some of Lewis' friends in terms of the phenomenon of projection.8 Anyone who has presented papers at philosophy conferences knows that the process of criticism and revision is just how things get done in that discipline. At such meetings, typically a presenter presents a paper, and a commentator offers a critical response. Sometimes the critical response is completely devastating to the arguments of the presenter. But more frequently the commentator finds difficulties with the paper that force the presenter to revise and strengthen his arguments. And this seems to have been exactly what happened in the Lewis-Anscombe controversy.

Did C. S. Lewis Lose His Apologetics in A Grief Observed?

Another dubious biographical interpretation has been given credence by the widely popular play and film Shadowlands, in which Lewis is depicted as abandoning his Christian confidence in the face of the tragedy of his wife's death, maintaining his religious beliefs only through a leap of faith. Quite coincidentally, Beversluis, whose book came out about the same time as the screenplay, maintains that Lewis abandoned his apologetic stance in the course of his grief experience, and that to retain his faith he defended his faith on grounds completely alien to the grounds he had used in his other apologetic writings.

In particular, Beversluis maintains that Lewis implicitly shifts from a Platonist to an Ockhamist conception of God in the course of A Grief Observed, and thereby embraced a fideistic understanding of faith and reason, jettisoning his entire previous apologetic work. Platonism is defined by Beversluis as the view that God's goodness is to be understood in a way that is continuous with our normal conceptions of goodness. Lewis had previously argued that the conception of God as good made sense only if that concept of goodness was continuous with our own. But Beversluis maintains that in the course of his grief experience he accepted a view of God according to which God's actions are right just because it is God who performs them.9 On this latter view, which Beversluis calls Ockhamism, no one could complain if God, before the foundation of the world, had chosen a few people to be saved and all the rest to be punished everlastingly. That this is an affront to human reason, according to Ockhamism, only shows that (natural) human reason is fallen and part of our desperately wicked human nature. Indeed, it is not surprising to discover that Ockhamism is popular among Calvinists, including Calvin himself.

But a careful reading of A Grief Observed suggests that Lewis not only does not abandon his previous apologetic arguments but in fact reaffirms them, including his arguments against Ockhamism. Lewis in his youth had been an atheist who had rejected theism based on the argument from evil, and had even written poetry from that perspective.10 Echoing that earlier atheistic perspective, Lewis speaks of God in the early part of A Grief Observed as "a very absent help in trouble," a "Cosmic Sadist," and an "Eternal Vivisector."11

But it seems evident to me that Lewis' primary reason for coming to accept the good God of Christianity was not because he thought he had some overwhelmingly plausible explanation for all the evils in the world, but rather because alternative worldviews had even more serious difficulties with them, and his treatment of these issues in A Grief Observed reflects this outlook on the problem. In A Grief Observed, Lewis briefly considers three possible alternatives to belief in a God whose goodness is commensurate with the way in which we use the concept of goodness in our ordinary discourse about humans: naturalistic atheism, the cosmic sadist hypothesis and Ockhamism. All of these alternatives, however, are rejected in the course of A Grief Observed for much the same reasons that they are rejected in his apologetic writings. It is true that the book is primarily pastoral, addressed to the grieving believer (first and foremost Lewis himself) rather than to the skeptic. Nevertheless the book does contain some argumentation to show that his grief experience does not provide any reason to adopt some worldview other than a theism that includes a Platonistic conception of divine goodness.

Let us consider the situation with respect to naturalism or materialism. In his apologetics Lewis had criticized naturalism because he considered it inconsistent with the validity of reasoning. One aspect of the argument from reason is the idea that physical states cannot be true or false. It is presented in the essay "De Futilitate":

We are compelled to admit between the thoughts of a terrestrial astronomer and the behavior of matter several light-years away that particular relation we call truth. But this relation has no meaning at all if we try to make it exist between the matter of the star and the astronomer's brain, considered as a lump of matter. The brain may be in all sorts of relations to the star no doubt: it is in a spatial relation, and a time relation, and a quantitative relation. But to talk of one bit of matter being true of another seems to be nonsense. It might turn out to be the case that every atom in the universe thought, and thought truly, about every other. But that relation between any two atoms would be something quite distinct from the physical relations between them.12

In explaining why, in the face of his wife's death, Lewis does not believe she has ceased to exist, he echoes this argument, though he talks in terms of falsity rather than truth:

If H [Joy] "is not" then she never was. I mistook a cloud of atoms for a person. There aren't and never were, any people. Death only reveals the vacuity that was always there. What we call the living are simply those who have not yet been unmasked. All equally bankrupt, but some not yet declared. But this must be nonsense; vacuity revealed to whom? bankruptcy declared to whom? To other boxes of fireworks or clouds of atoms. I will never believe—more strictly I can't believe—that one set of physical events could be, or make, a mistake about other sets.13

In other words, Lewis considered materialism as self-refuting, as he always had before. Now the argument he presents is hardly designed to persuade a skeptical philosopher like Anscombe or Beversluis, but it is a reaffirmation of his previous arguments.

Another counterposition he considered is the possibility that God is not a good being but a cosmic sadist, and he finds this idea also unacceptable. Beversluis explains this by saying that "The shift [from Platonism to Ockhamism] occurs when Lewis begins to suspect that the hypothesis of the Cosmic Sadist is too anthropomorphic. According to such a view, God is like the man who tortures his cats, and that is unbearable."14 But for Lewis, such a view of God is not simply unbearable; it is nonsensical. In Mere Christianity Lewis had considered the view he called dualism, according to which there is a good being and an evil being, both coeternal, who struggle for the control of the universe. Although Lewis finds this to be, "next to Christianity, the manliest and most sensible creed on the market," he finds it open to serious criticisms. These criticisms are precisely the criticisms he levels at the hypothesis of the cosmic sadist in A Grief Observed. Consider the following passage from Mere Christianity:

If Dualism is true, then the bad Power must be a being who likes badness for its own sake. But in reality we have no experience of anyone liking badness just because it is bad. The nearest we can get to it is in cruelty. But in real life people are cruel because they have a sexual perversion, which makes cruelty cause a sensual pleasure in them, or else for the sake of something they are going to get out of it—money, or power, or safety. But pleasure, money, power, and safety are all, as far as they go, good things. The badness consists in pursuing them by the wrong method, or in the wrong way, or too much. I do not mean, of course, that the people who do this are not desperately wicked. I do mean that wickedness, when you examine it, turns out to be the pursuit of some good in the wrong way. You can be good for the mere sake of goodness, you cannot be bad for the mere sake of badness. … In other words badness cannot succeed even in being bad in the same way in which goodness is good. Goodness is, so to speak, itself: badness is only spoiled goodness. And there must be something good first before it can be spoiled.15

A moment's reflection will reveal that this is not only an argument against dualism, but also against the doctrine of the cosmic sadist. The idea that the creator of the universe might be evil is not plausible since evil cannot exist on its own, but is always a perversion of good. And this apologetic argument is what Lewis uses to respond to the thesis of the cosmic sadist in A Grief Observed:

But the picture I was building up last night is simply the picture of a man like S.C.—who used to sit next to me at dinner and tell me what he'd been doing to the cats that afternoon. Now a being like S.C., however glorified, couldn't invent or create or govern anything. He would set traps and try to bait them. But he'd never have thought of baits like love, or laughter, or daffodils, or a frosty sunset. He make a universe? He couldn't make a joke, or a bow, or an apology, or a friend.16

Finally, and most critically, Lewis considers the possibility of replacing his good-in-our-sense God hypothesis with Ockhamism, the view that God's sadistic conduct is in the final analysis justified, because we are so fallen and depraved that our ideas of goodness simply do not count. According to Ockhamism, actions and commands are right because God does them; according to Platonism God does what God does because it is right. If God were to announce a new, reversed set of Ten Commandments, which commanded adultery, homicide and theft, an Ockhamist would say that these actions would be right because they were divinely commanded. In previous writings Lewis had harshly criticized Ockhamism. In The Problem of Pain he wrote:

It has sometimes been asked whether God commands certain things because they are right, or whether certain things are right because God commands them. With Hooker, and against Dr. Johnson, I emphatically embrace the first alternative. The second might lead to the abominable conclusion (reached, I think, by Paley) that charity is good only because God arbitrarily commanded it—that He might equally well have commanded us to hate Him and one another and that hatred would then have been right. I believe, on the contrary, that "they err who think that of the will of God to do this or that there is no reason besides his will."17

Beversluis maintains that Lewis accepts Ockhamism in A Grief Observed. But actually, that is the very book in which Lewis presents his most penetrating critique of Ockhamism. In most of his writings, Lewis is content to point out the morally disastrous consequences of Ockhamism. In A Grief Observed, Lewis points out that Ockhamism has disastrous consequences for knowledge as well:

And so what? This, for all practical (and speculative) purposes sponges God off the slate. The word good, as applied to him, becomes meaningless: like abracadabra. We have no motive for obeying Him. Not even fear. It is true that we have His threats and promises. But why should we believe them? If cruelty is from His point of view "good," telling lies may be "good" too. Even if they are true, what then? If His ideas of good are so very different from ours, what He calls "Heaven" might well be what we should call Hell, and vice versa. Finally, if reality at its very root is so meaningless to us—or, putting it the other way round, if we are such total imbeciles—what is the point of trying to think either about God or about anything else? The knot comes undone when you try to pull it tight.18

René Descartes, in order to raise skeptical doubts about even our firmest certainties, imagined that we might be under the influence of an evil demon, and more modern philosophers have speculated about the possibility of our being brains in vats. The Ockhamist hypothesis (and surely the cosmic sadist hypothesis as well) is as epistemically damaging as the brain in the vat hypothesis. On these hypotheses we will believe the truth only if a powerful being whose motives are either wicked or incomprehensible chooses that we should believe the truth, and not otherwise. These theses are, on Lewis' view, equally as self-refuting as is naturalism. If they are true then no one could know this, or anything else.

Of course, the above discussion does not demonstrate that these arguments are good ones, but only that Lewis continued to employ them throughout his career, even in the face of personal tragedy. I have been at some pains to show the continuity between Lewis' apologetic writings and A Grief Observed, a continuity that, so far as I can tell, has gone unnoticed by many commentators and denied outright by others. It is true that Lewis experienced a dark night of doubt and came to believe that his faith had been unreal. But what does that mean? It is my contention that it had absolutely nothing to do with the grounds of his intellectual assent to Christian theism. Rather, faith in this context should be understood as trust. The intellectual foundations of his faith were not shaken by his grief experience, as he reminded himself as he wrote the notebooks that became A Grief Observed. Nevertheless he experienced doubt. After reaching the conclusion that he had no real new grounds for abandoning his faith in favor of any counterposition, he still had to deal with his loss and understand why these doubts had arisen. And in doing so he drew the conclusion that he did not have the firm trust in God's wisdom that he had thought he had. But while this is a serious admission, it says nothing about the apologetic grounds on which he assented intellectually. The message of A Grief Observed is that those intellectual grounds remained what they had always been.

I think it is time for us to examine Lewis' thought fairly and honestly, not expecting either inerrancy or rank amateurism, but rather an incisive and powerful mind with many ideas that need to be pursued further. Moreover, we should resist, as far as possible, the "personal heresy" of focusing on Lewis himself rather than what he had to say.

Victor Reppert is adjunct professor of philosophy at Glendale Community College in Glendale, Arizona. This article is excerpted from C. S. Lewis's Dangerous Idea, by Victor Reppert, just published by InterVarsity Press. ©2003 by Victor Reppert. Used by permission of InterVarsity Press.

1. The most developed statement of the argument is found in C. S. Lewis, Miracles: A Preliminary Study (Macmillan, 1978), pp. 12-24.

2. G. E. M. Anscombe, Metaphysics and the Philosophy of Mind, vol. 2 of The Collected Papers of G. E. M. Anscombe (Univ. of Minn. Press, 1981), pp. 224-31.

3. My use of the term "Anscombe legend" is based on a comment by J. R. Lucas, who wrote, "The received version should be treated with some caution: Professor Mitchell, who attended all the meetings of the Socratic Club at that time, has no memory of the encounter. Oxford legends often owe more to the attitudes of those who report them than to the facts which allegedly they report" (J. R. Lucas, "The Restoration of Man: A Lecture Given in Durham on Thursday, October 22, 1992," J. R. Lucas Home Page, users.ox.ac.uk/~jlucas/lewis.html).

4. Humphrey Carpenter, The Inklings (London: Allen & Unwin, 1978), pp. 238-39.

5. A. N. Wilson, C. S. Lewis: A Biography (W. W. Norton, 1990).

6. John Beversluis, C. S. Lewis and the Search for Rational Religion (Eerdmans, 1985), p. 73; "Beyond the Double-Bolted Door," Christian History, Vol. 4, No. 3 (1985), pp. 28-31.

7. John Beversluis, "Surprised by Freud: A Critical Appraisal of A. N. Wilson's Biography of C. S. Lewis," Christianity and Literature, Vol. 41, No. 2 (1992), pp. 179-95.

8. Anscombe, Metaphysics, p. 10.

9. Beversluis, C. S. Lewis and the Search, pp. 102-3.

10. C. S. Lewis, Spirits in Bondage: A Cycle of Lyrics (Harcourt Brace Jovanovich, 1984).

11. C. S. Lewis, A Grief Observed (Bantam, 1976), see especially pp. 5, 35.

12. C. S. Lewis, "De Futilitate," in Christian Reflections (Eerdmans, 1967), pp. 63-64.

13. Lewis, A Grief Observed, pp. 8-9.

14. Beversluis, C. S. Lewis and the Search, p. 150.

15. Lewis, Mere Christianity, p. 49.

16. Lewis, A Grief Observed, pp. 139-61. The claim of this passage in A Grief Observed is precisely the same as the point made in the Mere Christianity passage: neither a wicked man like S. C. nor Satan could create anything; they can only pervert things that have already been created good.

17. C. S. Lewis, The Problem of Pain (Macmillan, 1962), p. 100.

18. Lewis, A Grief Observed, pp. 37-38.

Copyright © 2003 by the author or Christianity Today/Books & Culture magazine.
Click here for reprint information on Books & Culture.