C.S. Lewis and the Search for Rational Religion

A review of John Beversluis' book, C.S. Lewis and the Search for Rational Religion, by David Werther, Ph.D.

by John Beversluis on May 13, 2010


John Beversluis, C.S. Lewis and the Search for Rational Religion. Revised and Updated, Amherst, New York: Prometheus Books, 2007. 363 pp. Paperback, $21.98. ISBN 978-1-59102-531-3

In Mere Christianity C.S.Lewis asserts: "I am not asking anyone to accept Christianity if his best reasoning tells him the weight of evidence is against it.”  In C.S. Lewis and the Search for Rational Religion, John Beversluis explains why his best reasoning tells him to reject Lewis's arguments from desire, reason, morality, and, not least, the so-called lord-liar-lunatic trilemma. In his critiques, he not infrequently contends that Lewis is attacking a straw man (e.g. p. 228), offering a false dilemma (e.g. p. 105), or making some other argumentative error. If some readers sail by such things, Beversluis believes it is likely that they have been beguiled by Lewis's rhetoric. Take away the polished prose, analyze the arguments, and Lewis's case for Christianity loses its force (p. 22). Or so John Beversluis contends.

One of the great strengths of Beverluis's book is that it is full to the brim with detailed arguments, and evidences a close reading of the vast Lewis corpus and secondary literature. Indeed Beversluis intends his work not just as a critical examination of Lewis's apologetic but also of "that cadre of expositors, popular apologists, and philosophers who continue to be inspired by him and his books" (p. 11). In this review, I cannot begin to address all of the objections Beversluis raises. Instead, I will focus on on seven fundamental ones, beginning each of my discussions with an italicized quotation from C.S. Lewis and the Search for Rational Religion.

The concept of God as the universal object of desire derives not from the Old or New Testament, but from Plato; and the biblical texts that contradict it are legion (p. 62).

In The Pilgrim's Regress: An Allegorical Apology for Christianity, Reason and Romanticism, the sermon "The Weight of Glory," Mere Christianity, and Surprised by Joy, Lewis argues that we have a desire, "joy," that finds its satisfaction in a transcendent object. Among other things, Beversluis argues that Lewis's  appeal to desire is ethnocentric and outdated (p. 53), based on a weak analogy (p. 44), fails to take into account the propositional content of desires (pp. 48 ff.) and requires the reconciliation of two incompatible viewpoints (pp. 57 ff.): the Socratic tradition, holding that humans desire the good and err through ignorance, and the biblical view that humans knowingly spurn the highest good/God. This last objection is especially important for, if right, Lewis's argument is "shipwrecked . . . on theological grounds . . . " (p. 64).

Beversluis has no trouble finding Pauline proof texts to make his case for the alleged incompatibilty of Jerusalem and Athens. He rightly points to passages such as Romans 1:19-25 and Romans 3:10-11, scriptural data any Christian view of desire needs to incorporate. Conspicuous by their absence are two biblical passages that, prima facie, provide significant support for Lewis's view: Paul's speech to the Athenians (Acts 17:22-31), and the book of Ecclesiastes. Though humans are alienated from God and knowingly act wrongly, it does not follow that we do not and cannot seek God. What follows is that, left to our own devices, we could not seek God. But scripture teaches that God has not left us to our own devices. The Holy Spirit convicts us of sin and, according to the Wesleyan tradition, God extends prevenient grace to all persons. So long as we allow that our desire for God is always a response to God's reaching out to us, there is no incompatibility between humans being both alienated from God and longing for Him.

The legitimate and clarifying process of disambiguating terms is very different from the illegitimate and obfuscating process of redefining them (p. 229).

In The Problem of Pain, Lewis argues that the existence of pain (evil) is compatible with God's existence. To do so he examines two divine attributes: omnipotence and omnibenevolence. Following Aquinas, Lewis argues that omnipotence only pertains to doing what is free from contradiction. If humans have libertarian freedom (to use the contemporary designation for the type of creaturely freedom Lewis has in mind) and can act rightly only when and where wrongdoing is a possibility, then not even an all-powerful being can create a world with free creatures, who cannot possibly act wrongly

With regard to omnibenevolence, Lewis looks to scripture to clarify the concept of goodness. Beversluis argues that Lewis's clarification amounts to a significant departure from our ordinary understanding of goodness, though, in reference to the earlier edition of his work, he allows, "I no longer think that Lewis's view of God's goodness is quite as devoid of the recognizable goodness as I originally claimed" (p. 263). Even so, he still thinks ". . . he [Lewis] stretches the meanings of 'good' and 'love' almost to the breaking point" (p. 263). As such, even if Lewis shows that pain is compatible with the existence of a perfectly good being in his clarified sense of goodness, that is for all intents and purposes an irrelevant apologetic exercise. What is at issue is not whether the existence of evil is compatible with the existence of a perfectly good being, in Lewis's sense of goodness, but whether evil's existence is compatible with the existence of a perfectly good being, where goodness is construed more closely to its ordinary language usage.  

Pace Beversluis, insofar as the problem of evil is an internal problem within Christianity, a question of its own coherence, Christians should draw upon the resources of Scripture to clarify their understanding of goodness. Such clarification would go awry if the Biblical data was subjected to ad hoc readings or interpreted in such a way that scripture contradicts general revelation. I don't see that Lewis runs into trouble either way. But suppose worse comes to worst and Lewis’s understanding of goodness turns out to be incoherent. Even if this were so, what's left of Lewis's position would be quite impressive as it foreshadows Alvin Plantinga's free-will defense, a classic in analytic philosophy of religion. Lewis recognizes, as Plantinga did years later and with considerably more rigor, that if it is possible that creatures have libertarian freedom, then not even an omnipotent being can make them act rightly.  That alone is sufficient to show that, contrary to the logical problem of evil, the existence of evil is not necessarily incompatible with God’s existence. This is the nub of the free will defense and Lewis was right about that, even if his reading of "goodness" turns out to be problematic.  N B  Beversluis rejects this key claim (p. 242) but does not argue against it in this volume, as he holds he that “Lewis’s argument can be disposed of on purely linguistic grounds . . . “ (p. 243).  I leave an assessment of that “disposal” as an exercise for the reader.

Lewis defends God for having created the world as he did by calling attention to the 'danger signals' and 'warnings' that the pain fibers in our nerves are designed to transmit and by claiming that these signals would be necessary even in a perfect world. But he is clearly wrong. . . . he [God] could have eliminated the problem of pain and suffering in advance by creating human beings without the capacity to feel pain. . . . Or he could have created them with a much higher pain threshold that would have greatly minimized the degree of pain they are presently capable of experiencing. The resulting creature would have been very different from us, but its creation would not have involved any self-contradictory tasks (p. 248).

We can all imagine worlds without AIDS, cancer, child pornography, and genocide. So long as creating them is in the range of divine omnipotence we might think that it would be morally impermissible for God to create a world like ours, susceptible to such evils.  But why is that? Does God have a duty to do what is best, to create a world with the most happiness and the least pain? Not if the notion of the best possible world is like that of the highest possible number. Any world we take to be the best can always be "bested" by increasing its happiness (e.g. adding another creature to enjoy the beatific vision), just as no matter how high we count, we can always add one and get a higher number. But then, there cannot be a best possible world, a fortiori, there cannot be a divine duty to create it.

Even so, we might still ask "Given that we can imagine worlds without cancer and the like, doesn't it follow that evils of our world are incompatible with God’s goodness?” Not at all. No doubt there are some worlds which God could not create, ones where every creature rebels and no creature repents, or worlds where the evils of rebellion outweigh the goods of redemption, but it is not obvious that our world is one of them.  Showing that there are worlds better than ours does not demonstrate that a perfectly good being could not have created ours.

. . . having denied that God's goodness is the result of his conformity to the moral law or some independent standard but rather rooted in his 'essential nature,' Lewis cannot articulate what it means to say that God is good in this sense without making 'good' indefinable, unknowable, inexpressible, and arbitrary as it is for the Ockhamism he is determined to avoid (pp. 270-1).

Beversluis notes that in the essay, "The Poison of Subjectivism," Lewis argues, contrary to the Euthyphro dilemma (either something is right because God declares it so, or God declares it so because it is right), that the moral law is not either independent of God, or created by Him. And, that is a good thing for, on the one hand, if the moral law were created by God, we would have divine voluntarism/Ockhamism, and, on the other hand, if the moral law were independent of God, God would be in some sense be subservient to it: God would need to obey some external standard. After rejecting these alternatives, Lewis suggests that "God is not merely good, but goodness; goodness is not merely divine, but God."    

Beversluis believes that, by rejecting the view that goodness is a property God possesses, Lewis's claims about God's goodness leave us in the dark.

Unless there is something, some discernable and, in principle, expressible property that God embodies or manifests that constitutes his goodness and differentiates him from an evil God (or a God who is 'beyond' good and evil), Lewis's claims about his goodness are as vacuous as the Ockhamist's claims about the goodness of his commands. Lacking an elucidation of that property, to say 'God's goodness is expressive of his very nature, which is goodness itself' is to say nothing at all (p. 270).

Beversluis is right to note that Lewis’s identification of God and goodness rules out a deeper elucidation of goodness. But, that’s not a problem, for, as Mark Linville pointed out in conversation, that’s just what it means to offer an ultimate account of goodness. Further adapting an example of William Alston’s Lewis’s identification of God and goodness, need no more leave us in the dark regarding goodness, than the identification of water with H2O need have robbed our ancestors of any significant understanding of water.

Like Russell, Hume thinks that moral judgments are based on feelings-or as he prefers, on "sentiments;" unlike Russell, however, he does not mean the sentiments of the individual person who makes them. Lewis overlooks this when he insists that if our sense of values is a "purely human" sense, we can neither use it as the ground for what are meant to be "serious criticisms of the nature of things" nor "continue to attach any importance" to the efforts we make toward realizing our ideas of value, such as making sacrifices for the good of posterity. Hume disagrees. According to him, morality is a completely human enterprise. It is based on dispositions that have evolved over a long period of time, have taken deep root in human nature, and are all but universal"  (p. 84).

Beversluis accepts Lewis's criticisms of individualistic subjectivist accounts of the Russellian sort. However, he holds that Lewis neglects to consider Humean subjectivism, in which morality is grounded in nearly universal human sentiments. And, Humean subjectivism has the resources for both offering serious moral critique and undergirding the significance of moral efforts, two roles Lewis thinks subjectivism cannot play.

Pace Beversluis, switching from Russellian to Humean subjectivism, is not sufficient for escaping Lewis's critique. To see why this is not so, assume, for the sake of the argument, that there are nearly universal moral sentiments. Descriptive claims about these sentiments cannot be the basis of moral criticism because that requires value judgments. And, such prescriptive claims cannot be derived from purely descriptive claims about human sentiments. Here Lewis's comments in The Problem of Pain about attempting to infer obligation from first-person reports are pertinent: "You can shuffle 'I want' and 'I am forced' and 'I shall be well advised' . . . as long as you please without getting out of them the slightest hint of 'ought' and 'ought not.’"    For the same reason, switching from "I want" (cf. Russellian subjectivism) to "We want" (cf. Humean subjectivism) cannot get us any closer to the prescriptive judgments we need to offer moral critigues. Nor, given their “iffiness,” can prescriptive expressions of these Humean sentiments (e.g., "We ought to care for our parents, siblings, and children") undergird our moral strivings. As Darwin notes in The Descent of Man and Selection in Relation to Sex, ‘If . . . men were reared under precisely the same conditions as hive-bees, there can hardly be a doubt that our unmarried females would, like the worker-bees, think it a sacred duty to kill their brothers, and mothers would strive to kill their fertile daughters, and no one would think of interfering.’ (The quotation is from Mark Linville’s excellent article, “The Poverty of Moral Naturalism” in Contending With Chrisitanity’s Critics ed. Paul Copan and William Lane Craig [Nashville: Broadman and Holman Academic, 2009]).

. . . all four arguments [i.e. Lewis's dilemmas and trilemma/so-called "lord-liar-lunatic] are open to other objections [i.e. other than the incompleteness and ambiguity]. One of the most fundamental is that they all uncritically assume that the synoptic Gospels are historically reliable sources that accurately report what Jesus saidnarratives that preserve his very words (or a very close approximation) rather than narratives that incorporate later recollections, interpolations, embellishments, fictionalizations, and ascriptions of deity (p. 116).

Commenting on the so-called Trilemma, philosopher-theologian John Hick says that those who endorse Lewis's claim that 'someone claiming to be God must be either mad, bad, or God . . . continue to be unacquainted with modern biblical study,' which does not uncritically assume that the synoptic Gospels are accurate reports of what Jesus said (p.117).

Stephen T. Davis is a notable exception to the claims of John Beversluis and John Hick regarding advocates of the lord-liar-lunatic argument. In recent years Davis carefully developed the trilemma, with an eye to radical biblical criticism. Here is how he describes his approach.


I will present my argument in two stages. The first will presuppose the basic correctness of the methods and conclusions of some of the most radical of biblical critics. Its aim is to open the door to the possibility [Davis's emphasis] of showing, even on the methods of people like Bultmann, Perrin, and members of the Jesus Seminar that Jesus implicitly taught his own divinity. The second stage (which consists of five subarguments) will try to confirm the point that Jesus actually did [Davis's emphasis] this very thing. At this second stage, I will continue to eschew any naive or ahistorical view of the Gospels, but I will no longer consider myself limited by the views of the radical critics." (“Was Jesus Mad, Bad, or God?” in The Incarnation  An Interdisciplinary Symposium on the Incarnation of the Son of God ed. Stephen T. Davis, Daniel Kendall, SJ, and Gerald Collins, SJ [Oxford University Press, 2002]) 

Beversluis briefly refers to Davis's work in a single footnote (note 22 page 334), but otherwise does not mention it. Perhaps that is because Davis develops the trilemma to argue for a more modest conclusion, establishing “the rationality of belief in the incarnation of Jesus.” Whatever the case may be, it is a shame that Beversluis does not address Davis’s development of the trilemma, for, in passing over it, he fails to engage one the most (if not the most) powerful and sophisticated versions of the argument. Readers interested in the trilemma would do well to begin with Davis's work, and consider Beverluis's discussion of less sophisticated versions of the argument in light of it.

We can now dispose of the Lunatic or Fiend Dilemma once and for all. Since the proposition 'It is not impossible for someone to mistakenly claim to be God, and also be a great moral teacher" is false, it follows that . . . the Lord, Lunatic, or Fiend Trilemma is also blocked. . . (p. 135).

C.S. Lewis famously argued that Jesus could not have been just a great moral teacher. If he were not divine, as he claimed to be, then he would  have been either a deceiver, a liar, or deceived, a lunatic. But since neither of these alternatives is plausible, the only option left is divinity.           

Not so, argues Beversluis. Jesus might have sincerely but wrongly believed he was divine and still been a great moral teacher. If so, Lewis's  lord, liar and lunatic alternatives would not be exclusive and exhaustive. And so, we would not be compelled to accept Jesus’ lordship/divinity, once the liar and lunatic alternatives had been ruled out. In response to Beversluis, David Horner argues that, had Jesus been wrong about his divine identity, he could not have been a great moral teacher. On this point, Horner’s comparison of Jesus with Martin Luther King is instructive and worth quoting at length.

We consider Dr. King a GMT [great moral teacher], despite his lack of complete moral integrity, partly because he never claimed to possess the latter, and partly because there was coherence between what he did claim and how he lived. He uttered profound truths about liberty and racial equality, and he lived consistently and with integrity with respect to them, to the point of being jailed, beaten, and ultimately killed. I dare say, however, that if his central message had been the importance of sexual fidelity (or if he had turned out that he was actually a secret informer for the Ku Klux Klan), he would not in fact today be considered a GMT – no matter how exalted his teaching had been in other respects. The coherence between content and character is especially important when we specifically consider the teaching of Jesus, since his central claims – unlike those of Dr. King – had to do precisely with his own identity, including his own moral character, which he portrayed as sinless. The comparison between Jesus and Martin Luther King, I suggest, only illumines the radical uniqueness of Jesus' claims for himself. The differences between the two are most striking. What is required for Jesus to qualify as a GMT is coherence between his character and the content of his teaching-and this would in fact be possible only for one who was morally perfect.  (““AUT DEUS AUT MALUS HOMO: A Defense of C. S. Lewis’s Argument from Reason” in C.S. Lewis as Philosopher ed. David Baggett, Gary R. Habermas and Jerry L. Walls, [Downer Grove: InterVarsity Press, 2008])

If being a great moral teacher rules out mistaken claims of moral perfection a not inconsequential moral matter Beversluis has yet to dispose of the trilemma. If these reflections on seven of Beversluis's major objections are on target, then his critique is not as damaging as he thinks. That said, a couple of concluding observations are in order. First, no brief review can begin to do justice to all Beversluis' arguments. He rightly states that Lewis has not been well served by overly enthusiastic and uncritical acceptance of his work. C.S. Lewis and the Search for Rational Religion provides a much-needed corrective, even if it sometimes goes too far in the other direction.  Second, Lewis's case for Christianity does not survive Beversluis's critical discussion unscathed. At the very least, Beversluis shows that a good deal of fine tuning and elaboration is needed. And that is not too surprising. In philosophical apologetics, Lewis was a strategist, not a tactician. Tacticians are needed, but strategists come first, and that is where Lewis comes in and makes his greatest contribution.

David Werther, Ph.D., Faculty Associate
University of Wisconsin-Madison

May 2010

Tags: review, beversluis, book, c.s. lewis, search, rational, religion, david, werther

Previous Page