1 ‘Pure Love,’ in The Virtue of Faith (New York: Oxford University Press 1986), 188

2 Divine Hiddenness and Human Reason (Ithaca, NY: Cornell University Press 1993), 26

3 ‘The Impact of Scriptural Studies on the Content of Catholic Belief,’ in Stump, Eleonore and Flint, Thomas P. eds., Hermes and Athena: Biblical Exegesis and Philosophical Theology (Notre Dame, IN: University of Notre Dame Press 1993), 9Google Scholar. Dummett makes this point in a context completely different from mine.

4 I have two worries about 1'. First, in a certain frame of mind, one might read 1’ in such a way that it implies that God must have a reason to permit S -as opposed to anyone else — inculpably to lack theistic belief at t- rather than at some other time. Or to put it slightly differently, one might be tempted to read ‘God must have overriding reason to permit her [S] to fail to have theistic belief at t’ in such a way that it would not be enough were God to have a perfectly general reason to permit inculpable nonbelief, a reason that picks out neither S nor t specifically. My worry is that the proponent of 1’ might yield to the temptation just described. If I had the space, I would explain why this temptation should be resisted. However, I only have space to refer the tempted to Peter van Inwagen’ s essays ‘The Place of Chance in a World Sustained by God,’ especially 50-65; ‘The Problem of Evil, the Problem of Air, and the Problem of Silence,’ 77, n. 11; ‘The Magnitude, Duration and Distribution of Evil: A Theodicy,’ 103-4. Page references are to these works as they are collected in Inwagen's, van God, Knowledge and Mystery: Essays in Philosophical Theology (Ithaca, NY: Cornell University Press 1995)CrossRefGoogle Scholar. See also his ‘Reflections on the Chapters by Draper, Russell and Gale,’ in Howard-Snyder, Daniel ed., The Evidential Argument from Evil (Bloomington, IN: Indiana University Press 1996), 234-5Google Scholar.

5 ‘William Alston on the Problem of Evil,’ in Senor, Thomas ed., The Rationality of Belief and the Plurality of Faiths (Ithaca, NY: Cornell University Press 1994)Google Scholar

6 See, e.g., many of the essays in The Evidential Argument from Evil.

7 Experience, Explanation and Faith (London: Routledge and Kegan Paul 1984), 238

8 Much of section I relies on J.L. Schellenberg's excellent work noted above. I have, however, veered from his version of the argument at points that I think are worth veering from. See my review of his book in Mind 104 (1995). What I have to say in the sequel applies, mutatis mutandis, to his particular version of the argument.

9 For example, we might object to the move from 1 to 1’ in the following fashion: ‘Even if we have no direct voluntary control over belief, we still have indirect voluntary control; therefore, we would be rightly held culpable for not believing that God exists even if we did not have reasonable grounds for believing.’ By way of reply, note that the premise is false since even if we can do things that will drastically increase the likelihood of our believing a proposition, it does not follow that we have control over our believing it. At best, we can put ourselves in a better position to believe it, but after we've done that, it is not up to us what happens. We may or we may not believe the proposition, depending in typical cases on what we have to go on. As for the inference, even if we have indirect voluntary control over our believings, it follows that we would culpably fail to believe that God exists whilst lacking reasonable grounds only if there is nothing epistemically defective about believing whilst lacking reasonable grounds. But certainly there is something epistemically defective about that. If I'm right on this point, perhaps we can affirm the move from 1 to 1’ without even raising the issue of doxastic control. For surely, provided that one does not culpably fail in some other relevant respect, one does not culpably fail to believe that God exists unless one has reasonable grounds to believe He exists.

10 See Schellenberg, 65-9.

11 Schellenberg, 84-5

12 We need to think of goods fairly broadly here. The avoidance of bad states of affairs and the following of deontological prescriptions can count as types of goods.

13 Pascal, Butler, Kierkegaard, Hick, and Swinburne (among others) have answered this question in ways importantly different from the way in which I wish to answer it. My answer, however, is most like Hick's.

14 One might object: ‘Of course we do. God could bring about the good in question by simply preventing anyone from ever becoming so twisted and ruined.’ There are well-known direct answers to this objection. I shall not give them. Rather, I shall simply rule the objection a foul, since it is not in accordance with the rules of the present discussion. The reason why it fails to accord with the rules is that it conceives of the Argument from Divine Hiddenness in a way that its proponents do not intend it, namely, as hanging on the argument from evil. Proponents of the Argument from Divine Hiddenness wish to develop and defend their argument independently of other worries about the human condition.

15 Audi, Robert describes such a person in ‘The Dimensions of Faith and the Demands of Reason,’ in Stump, Eleonore ed., Reasoned Faith (Ithaca, NY: Cornell University Press 1993)Google Scholar.

16 What follows in this and the next two paragraphs is inspired by Eleonore Stump, who credits St. Thomas Aquinas. See her ‘Faith and Goodness,’ in Vesey, G. ed., The Philosophy in Christianity (Cambridge: Cambridge University Press 1989)Google Scholar.

17 There are other possibilities here. For example, if the Jove with which I am disposed to love God would stem from a deep-seated fear of divine reprisal, then, upon coming to believe that He exists, I will love Him; but it would be far better if my love were not motivated by fear but by an attraction to God's goodness. This is not to say that one ought not to fear divine reprisal; rather, it is to say that it would be far better if one's Jove of God were not motivated by such fear. Consider an analogy. Even though parents prefer that their young children do not love them out of fear, they may well want them to fear being grounded. Reflection on what Christ called ‘the first and greatest commandment’ may well reveal other respects in which one's love of God may be more appropriate for its object. See Matthew 22: 34-40.

18 One might urge that, in these cases, her love would be defective if she knew that I had a strong second-order desire to improve the source of my love for her or to improve the love to which I was disposed. Naturally, if I had such a desire, I would have to be aware of these defects in myself and I would have to acknowledge them as such. I grant that it is much more plausible to think her love would be defective in this case were she to fail to reciprocate explicitly my love for her. And I am willing to say the same for God. If I am disposed to love God upon coming to believe that He exists, but I am so disposed for the wrong reason or the love to which I am disposed is unfitting, then, if I acknowledge these defects and strongly desire to be rid of them, a perfectly loving God would not have the sort of reason for remaining hidden that I have been sketching here. Of course, it is quite difficult to tell whether there are any such people. See Objection 4 and my reply to it below for more on this point.

19 On this issue, see Zagzebki's, Linda ‘Religious Luck,’ Faith and Philosophy 11 (1994)Google Scholar.

20 This sort of inference is frequently made in evidential arguments from evil. For a defense of the condition I lay down in the text, but in the context of discussing a certain evidential argument from evil, see my ‘The Argument from Inscrutable Evil,’ in The Evidential Argument from Evil.

21 See The Evidential Argument from Evil.

22 I have benefited from audiences at the Society of Christian Philosophers’ meeting at the American Philosophical Association, Pacific Division (1994), the Pacific Regional meeting of the Society of Christian Philosophers (1995), and the Tiffany Forum at Seattle Pacific University (1995). For release time to work on this paper during Fall Term 1994, I am grateful for a Faculty Research Grant from Seattle Pacific University. For comments on earlier drafts, I am indebted to William Alston, Phil Goggans, Mark Heller, Larry Lacy, C. Stephen Layman, Wesley Morriston, Alastair Norcross, Philip Quinn, Mark Walhout, Linda Zagzebski and, especially, Frances Howard-Snyder and William Rowe.