According to Steven Bo~r 1 Plantinga's well-known Free Will Defense establishes only that God may not be able to create a world in which persons are free but never make morally reprehensible choices. Plantinga's argument, according to Bo~r, does not show that the evil consequences of evil choices are compatible with theism since God could have created a world in which the evil consequences of evil choices would always be prevented by "coincidence miracles." And since the evil consequences of evil choices are the moral evils which appear to be most difficult to reconcile with theism, the Free Will Defense is irrelevant to the problem of moral evil in its most serious form.
Coughlan 2 and Dilley 3 have rejected Bo~r's proposal. My main aim in this short paper is to show that if Bo~r's proposal is developed, all of the objections from Coughlan and Dilley can be met.
Coughlan assumes that in the world which, Bo~r suggests, God could have created there would not be natural evil. Dilley contends that Coughlan is mistaken and that Bo6r is not obliged to argue that evils which have a source other than human free choices would not occur in the world which, Bo6r suggests, God could have created. According to Diiley in Bo~r's world there would be, in other words, earthquakes, illnesses, deaths, bereavements, pain, and so on.
Who is correct? There is in fact no correct answer to the question of whether or not there would be natural evil in Bo~r's world. Bo~r's point is just that you cannot get a justification of evil consequences of choices from the Free Will Defense. He does not show (a) that no theodicy provides a justification of those consequences, much less (b) that no justification of natural evils in general can be found. Bo~r also implies that he accepts (a) and (b), but that he does so is no part of his case for limiting the implications of the Free Will Defense.
Coughlan's reason for thinking that in Bo~r's world there would not be natural evil is as follows: Bo~r says that since there could be a world in which there are * I am grateful to Marcia Baron, Hugh Chandler and Michael Levine for critical comments on the content of this paper.