1 In his Wisdom to Doubt: A Justification of Religious Scepticism (Ithaca NY: Cornell University Press, 2007), John Schellenberg presents the general argument from divine hiddenness in ch. 9. As far as I can tell, he doesn't question this assumption until section 4, about 23 pages into the argument. In ch. 10, Schellenberg presents restricted versions of the argument. Much of what I say in this paper can be applied to those arguments, but the general argument will be the focus of our discussion.
2 Schellenberg J. L. ‘The atheist's free will offence’, International Journal for Philosophy of Religion, 56 (2004), 1–15; idem Wisdom to Doubt, ch. 12.
3 Notice the necessity operator in the front of this premise. Although I will not argue the point here, I think my arguments can be applied to show that the value of divine–creature relationships does not support DH1, even if Schellenberg were to completely remove the operator.
4 E.g. Aijaz Imran and Weidler Markus ‘Some critical reflections on the hiddenness argument’, International Journal for Philosophy of Religion, 61 (2007), 19–20; Poston Ted and Dougherty Trent ‘Divine hiddenness and the nature of belief’, Religious Studies, 43 (2007), 192–194.
5 Schellenberg Wisdom to Doubt, 204–206.
7 J. L. Schellenberg Divine Hiddenness and Human Reason (Ithaca NY: Cornell University Press, 2007) 21; idem Wisdom to Doubt, 199–200.
8 Idem, Divine Hiddenness and Human Reason, 18–21; cf. idem Wisdom to Doubt, 199–200.
9 This claim seems particularly reasonable on the plausible assumption that propositions like (a) ‘God knows p’ entail (b) ‘God knows that he knows that p’, and (c) ‘God has infallible warrant for his belief that p’.
10 Recall that we stipulated that ‘in all likelihood’ means ‘a probability equal to or less than.01’. Next note that ($1,000×.01)+($0×.99)=$10. If we were to subtract the cost of buying the ticket from $10, we would get the expected value of buying the ticket (according to the procedure I outlined above). Hence, if the cost of the ticket is more than $10, it will outweigh the chance to win the $1,000.
11 Schellenberg J. L. ‘Response to Howard-Snyder’, Canadian Journal of Philosophy, 26 (1996), 460.
12 As I mean ‘better relationship’, putting a subject in the relating position would ultimately lead to a better relationship than he otherwise would have had if the subject has a relationship and he would not have had one were he not put in the relating position.
13 Strictly speaking, SRT is incompatible with the mere possibility that relevant conjunction obtains, so it is certainly incompatible with the possibility, given the counterfactuals of freedom, that it does. Assuming that God would create a significant number of free creatures, it is certainly possible that the best God can do is to actualize a world in which the relevant conjunction is true. But Schellenberg can always weaken the modality in SRT from ‘necessary’ to ‘necessary, given the counterfactuals of freedom’. Since Schellenberg can weaken SRT in this way, I focus on SRT's incompatibility with it being possible, given the counterfactuals of freedom, that the conjunction is true in one of the creation-worthy worlds.
14 The wording of this conjunct assumes that if someone wouldn't relate to God, then God would know it. This assumption seems to follow from God's knowing the counterfactuals of freedom.
15 First, note the set of best (i.e. creation-worthy) worlds may be empty or include only one member. Second, a world is feasible iff it is one that God can bring about, given the counterfactuals of freedom.
16 Although we are assuming that Schellenberg's free-will offence fails, it provides the resources to show that the relevant conjunction would never obtain, for it claims that God wouldn't create people with free will and would simply design them to relate with God when they were put in the relating position. Hence, there never would be a subject who fails to relate to God, and so at least one conjunct of that conjunction will never obtain.
17 This conjunct is really a simplification. As long as we are not assuming Molinism, it would need to read something like ‘in all likelihood S's being put in the relating position would not otherwise benefit anyone's relationship with God and there is a good or evil that competes with the value of divine–human relationships’.
19 See Schellenberg J. L. ‘Hiddenness arguments revisited (II)’, Religious Studies, 41 (2005), 287–303, 299–230, and Wisdom to Doubt, 217.
20 Idem ‘Hiddenness arguments revisited (II),’ 299, emphasis original; this passage seems reproduced in his Wisdom to Doubt, 216.
21 Consider an analogous line of reasoning: inferring that x is a sufficient reason for y (at least partly) on the basis of x is a strong reason for y. For inferences of this kind to be successful, we must have some reason to believe that there are no stronger reasons for ~y. I will always have strong reason not to kill a human being, but this fact will not provide me with reason to believe I will always have sufficient reason not to do so, unless I can rule out there being other stronger reasons to kill a human being. This would require having some way of ruling out my needing, at some point in the future, to defend myself or my family from an evil attacker. (I think I can rule out that possibility because it seems statistically unlikely that I would have to defend myself or my family to the point of death.)
22 I say that accommodation must lead to an overall better world, but my argument can succeed using the weaker claim that accommodation must lead to a world that is at least equally good. For whether accommodation leads to a world that is at least as good also depends on the counterfactuals of freedom. Nonetheless, I endorse the stronger claim because, if accommodation only leads to an equally good world, it wouldn't be inappropriate for God to remain hidden in order to attain some good incompatible with putting a subject in the relating position. (On the other hand, it wouldn't be inappropriate for Him to accommodate that good and put the person in the relating position either.) And I take it that for that good to be disqualified as a reason to remain hidden, it would need to be inappropriate for God to act on that reason.
23 Indeed, whether the relationship would be overall better if the reason were accommodated itself depends on what the counterfactuals of freedom are. Suppose that God is trying to decide whether to refrain from putting Seeking Suzy in the relating position or to put her in the relating position and then to give her a dark night of the soul (assuming she doesn't resist the relationship). Given how Suzy would actually respond in the two different cases, it may be that she would, in the long run, have a better overall relationship if God refrains from putting her in the relating position.
24 For ORC to support the claim that, given the counterfactuals of freedom, there can't be any competing goods, we would also need to know that all potential outweighing reasons can be accommodated; however, this claim isn't obvious either.
25 J. L. Schellenberg ‘What hiddenness reveals’, in Daniel Howard-Snyder and Paul K. Moser (eds) Divine Hiddenness: New Essays (Cambridge: Cambridge University Press, 2002) 41, first emphasis is original, the second is mine; cf. idem ‘Hiddenness arguments revisited (I)’, Religious Studies, 41 (2005), 201–205, 207.
26 A number of people made important contributions to the development of this paper, especially John Schellenberg, who read multiple drafts, and the audience at Notre Dame's Center for Philosophy of Religion: Tom Flint, Alicia Finch, Jerome Gellman, Sam Newlands, Nick Trakakis, Mike Rea, and Kevin Sharpe. Also thanks to Bill Rowe, Mike Thune, Adam Green, Erik Hanson, Kurt Liebegott, and Paul Draper for their comments on earlier drafts of the paper. Finally, the Center for Philosophy of Religion is to be thanked for providing helpful resources (office space and good coffee).