I start from Mill's words about Mansel and the problem of evil. In this dispute Mansel has generally been thought to have come off worst. However, Mansel was clearly right to this extent: that what would make a man a good man would not be the same as what made God good. This is because, quite generally, what makes something good of its kind, where we can talk about goodness at all, varies with the kind. With Aristotle we must say: the criteria are fixed by the thing's ergon, or at least by something analogous. On this account, it would seem that God must be perfect, since as an intelligent agent He does supremely well what is in the nature of an intelligent agent to do. This seems to lead to the right solution to the problem of evil. But Mill also had something useful to contribute. He saw that there would be a consequent difficulty in regard to revelation. This difficulty seems to be severe and insufficiently discussed.

According to Swinburne, one way of dealing with the guilt that attaches to a morally bad action is satisfaction, consisting of repentance, apology, reparation, and penance. Thus, Christ's life and death make atonement for human sin by providing a reparation which human beings would otherwise be unable to pay. I argue that the nature of God's creative activity entails that human beings can by themselves make reparation for their sins, merely by apology. So there is no need for additional reparation, and the satisfaction theory of the atonement is otiose. Following an insight of Swinburne's, I argue that satisfaction is not sufficient for forgiveness, since satisfaction does not place the wronged party under any obligation to forgive the wrongdoer. Christ's death merits the forgiveness of those sins for which human beings have made satisfaction. It does this in virtue of a divine promise to reward Christ's meritorious life with the forgiveness of such human sin.

I argue that the idea of a freely embraced eternal destiny in hell is deeply incoherent and implies, quite apart from its incoherence, that we are free both to sin with impunity and to defeat God's justice forever.

A frequent objection to the fine-tuning argument has been that although certain necessary conditions for life were admittedly exceedingly improbable, still, the many possible alternative sets of conditions were all equally improbable, so that no special significance is to be attached to the realization of the conditions of life. Some authors, however, have rejected this objection as fallacious. The object of this paper is to state the objection to the fine-tuning argument in a more telling form than has been done hitherto, and to meet the charge of fallacy.

Just prior to The Critique of Pure Reason's examination of the various arguments for God's existence, Kant discusses the conceptual relationship between the idea of an ens realissimum and that of a necessary being. The purpose of this paper is to illustrate the extent to which this discussion informs his claim that the cosmological argument depends upon the ontological argument.

In two recent articles in this journal, David Basinger and Nathan Nobis raise objections to my characterization of infant suffering and the problem that it presents to theism. My main theses were that infant suffering to death is not ‘horrendous’ in the technical sense defined, and that a good God need only balance off rather than ‘defeat’ such suffering. Basinger, on the other hand, claims that some infant suffering should be considered horrendous, while Nobis suggests that such suffering must be defeated by God rather than merely balanced off. In this response I will briefly summarize my view and then respond to Basinger and Nobis.

In a recent article in this journal, Richard Gale and Alexander Pruss offer a new cosmological proof for the existence of God relying only on the Weak Principle of Sufficient Reason, W-PSR. We argue that their proof relies on applications of W-PSR that cannot be justified, and that our modal intuitions simply do not support W-PSR in the way Gale and Pruss take them to.