Evil and a Finite God: A Response to McGrath

Abstract

P.J. McGrath has recently challenged the standard claim that to escape the problem of evil one need only alter one's conception of God by limiting his power or his goodness. If we assume that God is infinitely good but not omnipotent, then God can scarcely be a proper object of worship. And if we assume that if God is omnipotent but limited in goodness, he becomes a moral monster. Either way evil remains a problem for theistic belief. I argue that McGrath fails to distinguish between the deductive and inductive problem of evil and between a limitation in God's "strength" and a limitation in God's "ability to act", and that once these distinctions are made, his argument fails. According to P.J. McGrath, almost all theologians and atheologians at least implicitly argue that to escape the problem of evil "one need only alter one's conception of God by limiting his power or his goodness". But this is wrong, he tells us. First, if we assume that God is infinitely good but not omnipotent, then God can "scarcely be a proper object of worship". For "some evils which formerly existed have been eliminated by human ingenuity" and "to say that [God] was unable to do so is to reduce his power to such an extent that it is difficult to see how he could any longer be regarded as divine". Second, if we assume that God is omnipotent but limited in goodness, he becomes "a moral monster". For only a being "more akin to an evil demon than to a deity" would "tolerate the sorts of evils which actually exist simply because of his lack of concern for the welfare of others". And, finally, if we assume God is limited in both goodness and power, we can avoid some of "the difficulties surrounding the second option". But this option is "even less acceptable than the first", for a being doubly limited is even less worthy of worship than one limited only in power. Thus, he concludes, "evil constitutes a problem for belief in even a scaled down version of deity"'! McGrath's conclusion is important, and his discussion has an initial ring of plausibility. But he fails to make two crucial distinctions which must be made in this context. OnGe made, his argument is greatly weakened.