“If God is all-powerful, then he must be able to abolish evil; if God is all-loving, then he must wish to abolish evil; but evil exists, therefore God cannot be both all-powerful and all-loving.” So runs the traditional statement of the problem of evil. Anthony Flew and John Mackie believed that the argument exposes the inner incoherence of theism; how can one harmonize this world with the traditional attributes of God? More recently, it has been treated as a good inductive argument against the existence of God: surely it is very unlikely that a God with the traditional attributes would have created a universe like this.

John Hick’s classic Evil and the God of Love , first published in 1968, set the contours for all subsequent discussion of the problem. He was responsible for identifying two different responses in the Christian tradition. The first and major response, expounded by Augustine, places all the blame for evil on fallen creatures (both human and angelic). Because of their disobedience the creation has been marred. On this view, evil serves no purpose in the plan of God. The second response is found in Irenaeus. Here evil and suffering are among God’s means for creating souls fit for heaven. Humanity was created immature; and life is a vale of soul-making. Hick is a passionate advocate of the second view.

Douglas Geivett, Associate Professor of Philosophy at Biola University in La Mirada, California, has set himself the task of explaining why Hick’s advocacy of an Irenaean theodicy is inappropriate, and how the Augustinian tradition can be affirmed. His book is scrupulously fair to Hick-a model of accurate reporting-and shows the ways in which Hick’s theodicy is of a piece with his opposition to natural theology (in the sense of proofs for God’s existence) and his belief in religious pluralism.

In responding to Hick, Geivett essentially takes up three themes. First, he draws attention to the significant role of natural theology in the Augustinian tradition. He points out that in De libero arbitrio voluntatis , Augustine is intent on proving the existence of God before formulating his theodicy. It follows that if God’s existence can be established on other grounds, the theodicy problem is significantly reduced. Unlike Hick, he believes that natural theology can succeed here, and he draws upon a significant range of contemporary arguments. His favorite seems to be the Kalam cosmological argument, much beloved by William Craig: given that cosmologists tend to believe it is likely that the universe had a start, and given that it is difficult to imagine a universe coming into being uncaused, then it is likely that God was the first cause. This argument is then combined with the anthropic principle (the universe must have properties that make inevitable the existence of intelligent life), providence, answered prayer, and morality to form a cumulative case for the existence of God. God, on these grounds, is more likely than not. In this sense, we start discussing the problem of evil with the significant advantage of knowing that God exists.

Geivett’s second theme points to the difficulties with Hick’s theodicy. The anthropology is inconsistent: it would be immoral of God to cause (and use) evil as a means to the creation of soul. Hick’s version of reincarnation (located in subsequent universes, not this one) and denial of hell undermine the real significance of human freedom-a freedom that can thwart the love of God for eternity.

The final theme stresses the coherence and achievement of a traditional Augustinian theodicy. God is not the direct cause of evil; humans were created innocent, but exercised a good will badly. Our purpose is worship; and hell awaits those who succeed in rebelling.

The case for linking natural theology with the problem of evil is sustained brilliantly. Moreover, one can supplement Geivett’s case here, for it seems clear that within our culture the problem of evil has assumed an ever greater significance as natural theology and the belief in life after death have been undermined. For the Christian tradition, the context for discussing evil has included certainty about God’s existence and the reality of life beyond the grave. As this context has disappeared, theodicy has become an increasingly insurmountable problem. Geivett is thus right to attempt to recover this context.

Hick, in a very interesting afterword, objects to the enterprise by citing the old claim that Kant and Hume demolished all the arguments for God’s existence, and claims that Geivett himself is not really interested in attaining a proof of God’s existence but only in illustrating that it is highly probable. Here Hick is buying into the unacceptable Cartesian standard for knowledge: that it must be established beyond any doubt. However, we now know that this is an impossible standard to meet. We cannot even have such certainty about the existence of other minds. And once we see this flaw in Hick’s objection, Geivett’s procedure seems entirely reasonable.

Hick is correct, however, in drawing attention to other areas of weakness in Geivett’s case, the most shocking being Geivett’s complete neglect of the Fall and its relation to natural suffering. For Augustine the Fall is central. According to him, God created an idyllic paradise with no suffering, death, or natural disasters. It was human disobedience that introduced these things into creation. Most educated people in the West no longer find that account plausible. The whole process of evolution involved the suffering and death of millions upon millions of creatures, within a context that included occasional earthquakes and floods. These facts are now part of the scientific data we must accept. And they strongly suggest that God, in the process of creation, built in the inevitability of suffering as a part of His method.

Geivett does not tackle this problem directly. When it comes to natural evil, he offers two alternatives that he considers compatible with an Augustinian theodicy. The first is Alvin Plantinga’s fallen angels, who cause earthquakes and floods. The second is that fixed laws are essential for freedom, and earthquakes are simply the inevitable side effects of these laws. It is worth noting that these two alternatives are contradictory: the first views earthquakes as violations of the natural order, while the second views them as an appropriate expression of the natural order. More fundamentally, the former is scientifically implausible-at what point in the process are we to postulate the evil action? And the second, although a much better suggestion, still leaves the mystery of why God does not intervene to mitigate the effects of the natural laws. Presumably, certain goods must emerge from the suffering caused by an earthquake to justify the nonintervention. If so, then the problem of theodicy remains.

Despite these problems, this is an elegantly written book and a valuable contribution to the discussion of theodicy.

Ian S. Markham is a lecturer at the University of Exeter and author of Plurality and Christian Ethics (Cambridge University Press).