William Hasker's The Triumph of God over Evil: Theodicy for a World of Suffering is a sustained attempt to make philosophical sense within a mainstream Christian theology of the justifying purpose of both evil and creation. Although it is hence philosophical in character, it is addressed, not to the sceptic about God's existence, but to someone who accepts the broad outlines of a mainstream Christian worldview and therefore to a traditionally Christian audience.

Chapter 1 identifies more specifically exactly what Hasker's project is. The idea is not merely to provide a defence to the problem of evil, i.e. a claim that is possibly true that shows the consistency of the existence of evil with the existence of a perfectly loving God. It is, rather, to provide a theodicy, i.e. a true claim that reconciles evil with God's existence and explains evil in this world. As suggested above, Hasker wishes to construct a theodicy drawing on the standard theologies of classical theism, the Trinity, the Incarnation, the Atonement, and open theism. As Hasker describes open theism, it is the view that God is affected by and responsive to human actions but does not know in advance all the actions that humans will freely choose.

Chapter 2 concerns the implications of the Holocaust for the Christian worldview. Hasker rejects the idea that the occurrence of the Holocaust implies that God is imperfect morally, a claim defended by John Roth and D. Z. Phillips. As Roth and Phillips see it, God permits evils that should not be permitted and rules the universe in a way that demands a response that is both loving and protesting. Hasker argues persuasively, among other things, that the assumption that such grave evils are gratuitous creates problems for the idea that we ought to worship God. Even so, Hasker acknowledges that an adequate Holocaust theodicy remains to be given.

Hasker turns in Chapter 3 to the logical problem of evil. He considers a number of responses, including the debate between Alvin Plantinga and J. L. Mackie on the free-will defence, concluding that Plantinga's claims that God creates a world that contains free will as a moral good, and all possible persons are transworld depraved suffice to show that God and evil are logically compatible. Hasker also considers the ‘concrete logical problem of evil’. The idea, as he describes it, is that a morally perfect God would give each person a life of great good but, while such a life would involve some evil, it could not involve horrendous evils because such evils cannot be compensated for or balanced off. Hasker responds that there is no reason to think that this is a logical truth about the nature of a morally perfect God.

This is a minor quibble but, although Hasker is correct in denying that this latter claim is a logical truth, there is no reason to think that it is a logical truth about the nature of a morally perfect God that God would prevent any evils. The logical problem of evil, as well as the solution, involves substantive moral claims: the idea that a morally perfect being would allow some evil if necessary to achieve a greater moral good is not a conceptual or logical truth; it is a substantive moral truth. To rebut fully the concrete logical problem of evil, substantive moral analysis is needed to address the claims that horrendous evils cannot be compensated for or balanced off.

Inasmuch as Hasker wishes to explain God's triumph over evil, it is not enough simply to rebut the logical problem of evil; accordingly, in Chapter 4, Hasker addresses the question of whether we can understand why God created this world, rather than nothing or another possible world. He argues that, first, there is no best possible world, rejecting Leibniz's famous claim that this is the best possible world. Second, he argues that if there is a best possible world, it is impossible for God to create it in advance because free creatures will invariably make choices that determine which possible world is actualized, except insofar as there are true counterfactuals of freedom that God can know in advance. Further, assuming open theism, God is moved and affected by what happens and hence creates because He can be enriched by what goes on, even while He has no need for creation.

In Chapter 5, Hasker attempts to provide a Christian theodicy of evil, considering and ultimately rejecting the view, defended by Henry Morris and quite differently by William Dembski, that natural evil is the result of human sin. He also evaluates the suggestion that suffering provides a believer with a way of experiencing God's love and participating in Christ's suffering – a gesture expressing the most perfect love of God of which we are capable. Hasker goes on to argue that there are structural features of this world that are good for God to create and that natural evil is the consequence of some of these very features. The disvalue of natural evil is outweighed by the value of these structural features.

Up to this point, the argument seems quite cogent; the logical problem of evil makes a claim that is simply too strong to be plausible and can be rebutted by a variety of different claims. Chapter 5 begins what I take to be the bulk of his theodicy, and here the claims seem more vulnerable to objections. Although Hasker intimates that he believes that a literal reading of Genesis is incorrect, it nonetheless creates a problem for his argument. If the idea that the first human beings existed in a physical world in which there was no death and no violence (all creatures were given, according to Genesis 1, only plants and fruits for food) and yet were free moral agents, then God could have created a world with the good structural features of this one without the disvalue of its natural evil. As we will see, producing a theodicy that draws on the resources of a mainstream Christian theology has its advantages, but it also has some significant disadvantages as well.

The question of whether it is plausible to think free will so valuable that God would allow great suffering to obtain that value is taken up in Chapter 6. Hasker asks us to consider the following situation: suppose that you are a prospective parent and are offered two options with respect to your child. First, ‘the child will, without any effort on your part, always and automatically do and be exactly what you want him or her to be’ (155) without feeling constrained or controlled, spontaneously making the choices you want him or her to make. Second, the child will develop in the normal fashion, with a will of his or her own, sometimes following your wishes, sometimes not – causing you both great joy and pain in the process. On Hasker's view, we would nearly all choose the latter option. If so, we are committed to the principle that the value of free will outweighs the suffering it produces.

Now there is, I want to be clear, much more to Hasker's sophisticated defence than I can discuss here, but the defence as a whole rests on this (and one other similar) example. I have challenged the claim that free will contributes moral value to the universe elsewhere in this journal on different grounds (‘The free-will defence: evil and the moral value of free will’, Religious Studies, 45 (2009), 395–415; ‘Plantinga's version of the free-will argument: the good and evil that free beings do’, 46 (2010), 21–39), but there seems to be a serious problem with this example. Notice that the joy and pain that is discussed in this example is the parents' and not the child's, and this cannot be ignored in exploring intuitions about free will.

Suppose that the difference between doing the right thing and doing the wrong thing is eternal bliss in heaven or eternal torment in hell – and hell is the probable fate of at least two-thirds of the world's population, as would be true if Christian exclusivism is true. (I should note that Hasker is an inclusivist – but this is not a view that is widely shared among traditional Christians, especially evangelical Christians.) It does not seem so obvious that most people would choose to let the child blunder his way through the world, where his mistakes might result in eternal torment. I certainly would not; in fact, I have argued elsewhere that, given Christian exclusivism and the traditional doctrine of hell, birth has to be regarded as a misfortune of sorts (‘Birth as a grave misfortune: the doctrines of hell, exclusivism, and salvific luck’, in Joel Buenting The Problem of Hell (Aldershot: Ashgate, 2010); ‘No harm, no foul: abortion and the implications of fetal innocence’, Faith and Philosophy, 19 (2002), 172–194).

In Chapter 7 Hasker takes on what I think is the strongest argument against the existence of God – the evidential problem of evil, the argument that there is more evil in the world than necessary to achieve any greater goods and is hence incompatible with the existence of a morally perfect God. The argument relies on ‘Rowe's Requirement’, which Hasker describes as follows: A perfect being ‘would prevent the occurrence of any intense suffering it could, unless it could not do so without thereby losing some greater good or permitting some evil equally bad or worse’ (177). Hasker argues that this principle is inconsistent with the principle that ‘it is an extremely important part of God's intention for human persons that they should place a high priority on fulfilling moral obligations and should assume major responsibility for the welfare of their fellow human beings’. The inconsistency arises, on Hasker's view, because if Rowe's Requirement is true, then people will not satisfy their moral obligations to help others in suffering because they will assume that their help would interfere with achieving the greater good that God intends to achieve by allowing that suffering.

This is a subtle and deservedly famous argument by a gifted philosopher, so I want to acknowledge at the outset that I am quite likely to be mistaken here. But I must confess that do not see any inconsistency here whatsoever. The claim that it is an extremely important part of God's intention that human beings obey obligations given to us through revelation and natural reason is perfectly consistent with Rowe's Requirement. The fact, if it is a fact, that human beings might be less likely to live up to their obligations if they believed Rowe's Requirement does not entail the falsity of the claim about the content of God's intention. God could have an intention that human beings perform certain acts that we chronically fail to perform. Indeed, that is precisely the tragic state of human beings in a fallen world.

But, equally importantly, it seems implausible to think that human beings would shy away from performing their obligations under such circumstances – at least not Christians. Scripture is absolutely clear that we have positive and negative obligations that are utterly unqualified by any perceived or unperceived consequences of not fulfilling those obligations; as Hasker observes elsewhere, mainstream Christian ethics seems obviously to reject consequentialism. Ignore those obligations, scripture makes clear, at the risk of eternal torment.

It is important to note that each of the concerns that I have raised are grounded in mainstream Christian theology. While Hasker, I think, believes that making use of the resources of Christian theology is an advantage in articulating a theodicy of evil, my view is that it is a profound disadvantage. Hasker assumes, much too optimistically on my view, that mainstream Christian theology obviously forms a coherent body of thought, but this is surely not obvious to any Christian who has ever grappled with the philosophical problems involved in these mainstream doctrines. Is there an uncontroversial account of how to reconcile God's perfect mercy and loving nature with an exclusivist doctrine of hell as eternal torment? Is there an uncontroversial account of the justice of Christian exclusivism? Is there an uncontroversial account of the coherence of the doctrine of the Trinity? Or of the necessity of the Atonement? Many philosophers of religion concern themselves primarily with general problems with theism, but the most important problems, I believe (as a committed Christian), arise in connection with the specific doctrines of Christianity. Assuming the coherence of these doctrines when there are so many unresolved philosophical issues seems to me to be a mistake – even apart from the fact that such a theodicy could never reach non-believers who, on mainstream assumptions, need that theodicy much more than Christians do.

Although I tentatively submit a few concerns about Hasker's argument here, the book is a remarkable one for a variety of reasons. It is extremely readable, frequently quite persuasive, explicates difficult arguments clearly, and is extremely creative. It is not only a worthwhile read for the professional philosopher and theologian; it would make a wonderful text in any class on the philosophy of religion. This is one more exceptional achievement in a career that is full of them, and Hasker deserves congratulation for this fine effort.