William Lane Craig and Walter Sinnot-Armstrong, God? A Debate Between a Christian and an Atheist. Oxford University Press, 2003.

During the past twenty years, while many apologists and philosophers have been debating (mostly among themselves) whether it is theologically acceptable or philosophically credible to argue for the existence of God, William Lane Craig has been on the road debating the existence of God with some of the leading unbelievers of our day. Through these encounters and his writings, both scholarly and more popular, Craig has significantly revived the ancient and venerable art of natural theology: the endeavor of employing rational arguments in support of the existence of a monotheistic God. How far can these arguments take us theologically toward the existence of the God of the Bible? Perhaps the proof is in the philosophical pudding. One of the best laboratories for answering this question is a rigorous philosophical debate—exactly what this short but weighty book offers.

This time Craig takes on Walter Sinnott-Armstrong, a well-established philosopher and a worthy opponent. The book is divided into two sections. Craig first presents the case for God, receives rebuttal by Sinnott-Armstrong, and then responds to the rebuttal. The order is reversed in the second section in which Sinnott-Armstrong begins by arguing against God. Although the book is introductory, both philosophers wade into deep waters and raise issues warranting further reflection. Instead of reviewing all the sparring, I will focus on a few strong points made by each philosopher.

Craig gives five short arguments for the existence of God instead of developing one argument in detail. This strategy allows him to develop a brief cumulative case argument that, if successful, secures several essential divine attributes instead of those derivable from only one argument. Not surprisingly, given his work on this subject, Craig begins with the kalam cosmological argument. This argument has a simple deductive form: (1) What ever begins to exist has a cause of its existence. (2) The universe began to exist. (3) Therefore, the universe has a cause. To make his case, Craig must explore the nature of infinity.

An actual infinite is a set of items that is literally limitless. This contrasts with a potential infinite, which is always finite but ever increasing. But the concept of the actual infinite generates contradictions. If you subtract all the odd numbers the set of natural numbers you are still left with an infinite number of numbers. Thus, infinity minus infinity equals infinity. But if you subtract all the numbers greater than two, you are left with three numbers. Now, infinity minus infinity equals three. Craig comments: “In both of these cases we have subtracted identical quantities from identical quantities and come up with contradictory answers.” The actual infinite is, therefore, rendered impossible. Consequently, the number of past events must be finite, and the world must have a beginning; we cannot affirm the world has always existed. Furthermore, the beginning of the universe is scientifically confirmed through Big Bang cosmology, which claims that the universe exploded into existence out of nothing about 15 billion years ago. Since it is absurd to think that the universe could pop into existence without a cause, Craig infers that the cause of the universe was a timeless “personal agent who freely chooses to create an effect in time without any prior determining conditions.”

Craig offers four more crisp arguments: the existence of God makes sense of (1) the fine-tuning of the universe, (2) the existence of objective moral values, and (3) the life, death, and resurrection of Jesus. He concludes by arguing that (4) “God can be immediately—and without argument—known and experienced as “properly basic.” He warns: “We mustn't so concentrate on the proofs for God that we fail to hear the inner voice of God speaking to our own heart.”

Sinnott-Armstrong responds in the tradition of David Hume by claiming that even if Craig's arguments succeed, he has not established the existence of an all-good, all-powerful, all-knowing, eternal, effective, and personal God. In fact, the creator, designer, and moral lawgiver may be different finite beings. Not surprisingly, he proceeds to give naturalistic explanations for all the data from which Craig derives theistic explanations. He responds to the kalam argument by claiming that the actual infinite is commonly employed in transfinite mathematics and poses no problems. He also disputes Craig's interpretation of the Big Bang by invoking other speculative possibilities that do not require an ex nihilo origination of the universe. He is particularly bothered by the notion that a timeless cause (God) can produce a change in time, the creation.

Craig rebuts Sinnott-Armstrong by defending his cumulative case approach, which collects several important theistic attributes from several different arguments. If favorable, these arguments establish a being who is a personal Creator, uncaused, eternal, changeless (at least before creation), immaterial, enormously powerful, inestimably intelligent, concerned with his creatures, good and loving, who exists by metaphysical necessity, is the God and Father of Jesus Christ, and is knowable. Craig, however, does not directly critique the claim that each argument may refer to different deities. But this charge is ill advised since these attributes are those historically stipulated by monotheism (an argument for polytheism would be quite different) and because one ought not multiply entities beyond what is required for a simple and sufficient explanation.

Craig seems especially to get the better of Sinnott-Armstrong on the origin of the universe. He responds that while actual infinites are allowed in transfinite mathematics, this is an abstract conceptual discourse that abides by certain artificial rules—a point not addressed in his initial argument. He refines his claim by arguing that the actual infinite cannot exist in space-time reality. (Craig usually hits this home by arguing that even if an actual infinite exists, it could not be traversed in time, since there would always be an infinite distance left to travel. But he does not make that precise point here.) Craig argues that the alternatives to the Big Bang presented by Sinnott-Armstrong are either outmoded or too avant-garde to displace the Big Bang and its metaphysical implications. Craig defends the idea that a timeless being can produce an act in time—an argument I have never found persuasive—but also admits the possibility that God existed in “undifferentiated metaphysical time” (the view of Alan Padgett) prior to the creation. In that case, God created within a condition of pure duration, not atemporality. However, this duration, existing without physical objects or laws, is not subject to any metric (time measurement); therefore, it does not suffer the fate of comprising an (impossible) actual infinite of discrete moments.

Sinnott-Armstrong's argument against the existence of God is threefold. He argues that some kinds of evil are incompatible with an all-good and all-powerful God, that an eternal God could not produce effects in the temporal world, and that the prevalence of unbelief does not comport with the existence of a God who supposedly wants people to know him. The latter is really a subset of the problem of evil (sometimes called “the epistemological problem of evil”), although Sinnott-Armstrong does not frame it that way. His first argument seems the most challenging.

In his response to Craig's moral argument for God, Sinnott-Armstrong (rather unconvincingly) affirms that moral propositions such as “rape is wrong” are true without further justification. This is pivotal to his argument against God, because without the existence of objective moral properties, the problem of evil cannot get off the ground. Sinnott-Armstrong builds a thorough argument from evil against God by critiquing each of ten possible strategies for justifying evils (such as the compensations of heaven, that suffering may produce better character, and so on). He limits his selection of evils to natural evils (such as infant deaths), so that the free will defense cannot directly address them. His arguments are nuanced, thorough, and impressive. They attempt to put a crushing burden of proof on the theist to show that all evils are adequately compensated by some good that could not occur without them. For Sinnott-Armstrong, if any evil is uncompensated or pointless, this counts decisively against the existence of an omnibenevolent and omnipotent God, who surely would justify any evil that God allows. Sinnott-Armstrong grants that some evils are compatible with God's existence, but not the kinds he discusses.

Craig retorts by arguing that a combination of possible justifications for evils may cover all of them, even natural evils affecting innocents. In this way he sketches a positive theodicy. However, he rightly notes that given our finitude, we cannot justify the reasons for every particular evil. Sinnott-Armstrong claims in his response that the mere possibility of these justifications offers a much less plausible argument than the notion that actual evils are simply inadequately compensated for.

The numb of the argument boils down to Craig's formulation: 1. If God exists, gratuitous suffering does not exist. 2. God exists. 3: Therefore, gratuitous suffering does not exist. Craig's strategies for God's possible justification of evils are merely speculative unless one has previously made a strong argument for a rich theism through natural theology. That is, background knowledge weighs crucially here. Since Sinnott-Armstrong denies the success of Craig's arguments for God, Craig's explanations for evil ring hollow and desperate to him. But if one takes Craig's overall, fivefold case for God to be strong (as I do), this would take the bite out of Sinnott-Armstrong's objections, since an infinitely intelligent, good, and powerful God would have a vast arsenal of reasons and strategies to employ with respect to justifying evils. Since Craig argues forcefully—if briefly—for the resurrection of Jesus as part of the cumulative case for God's existence, it might have served him well to invoke the resurrection as part of the solution to the problem of evil as well. If Jesus has been raised victorious over death and sin, the world is not without hope. Evil does not have the last word.

The debate over God will continue until the Last Day. Until then, God: A Debate Between a Christian and an Atheist, should help thoughtful inquiries begin to find their way concerning the Question of all questions.

Reviewed by Douglas Groothuis, Ph.D.
Professor of Philosophy
Denver Seminary
February 2005