Hiddenness and belief in God

As promised, I will regularly blog on some philosophy of religion topic, in keeping with the broad range of interests on NewApps. This one is on divine hiddenness, and is motivated by a nice reading group we are having at Oxford, led by Aku Visala, who does research on philosophy of religion and cognitive science. I am wondering here whether one of the crucial (in many formulations hidden) premises of the argument from hiddennesss is correct, namely whether having a relationship with God requires belief. 

Schellenberg's argument from divine hiddenness against theism says that

1. if God exists, he is perfectly loving
2. If a perfectly loving God exists, he wouldn't allow non culpable unbelief (e.g., atheists who are willing to believe, who have sincerely weighed the evidence and considered theism, but rejected it [Schellenberg pointed out to me that it's not only atheists but also, e.g., people in the distant past who have not been acquainted to theism are subject to nonculpable unbelief, see the update below]
3. However, there is nonculpable unbelief
4. therefore a perfectly loving God doesn't exist
5. There is no God

(2) is usually defended on the basis that a perfectly loving God would not withhold the benefits of a relationship with Him from people, which is the case for atheism. There are various counterarguments to this view, mainly directed against (2) (Some authors also take 3, namely that nonbelievers are always culpable, because evidence for theism is there for all to see, or they are not responding to their innate God sense (sensus divinitatis). However, an interesting question arises: Does one have to be a theist to have a meaningful relationship with God? Here there are a few plausible counterexamples (some of which mine, some of which from other sources.

Update: John Schellenberg e-mailed me to say that I omitted some relevant features of his hiddenness argument. First: he thinks it is not just atheism, but rather, any example of nonresistant nonbelief at all provides a basis for an argument to atheism -- even that of persons in the present or the distant past who lack a conception of God clear enough to make either theistic belief or disbelief a live option for them. Second, he finds it important to stress the conscious nature of belief (I quote from a handout of the St Thomas seminar he sent me): "An unsurpassable love, for many reasons, would seek meaningful conscious relationship with the beloved; it would always at least be open to sharing itself in such relationship."

• The tapping case: A nice example from Poston & Dougherty from their paper in 2007, Religious Studies (pp. 190-191), which I here quote in full: "Suppose that Jones – an unfortunate fellow – is locked in solitary confinement in a dark prison cell. Jones hears faint taps coming from the other side of his prison wall. The taps resemble the presence of another person willing to communicate, but it is not certain that there is another person in the other cell. Yet, Jones begins to tap back. Suppose this activity continues over a long period, and Jones can – with some effort – make sense of the taps as another person attempting to communicate with him. Suppose Jones’s credence (his degree of belief, rational confidence, or what have you) on the claim ‘there is another person in the cell beside me’ is 0.5. He seems to be discerning messages, but he realizes that it could just be in his head since the signs are ambiguous. Yet, given that the two persons are tapping back and forth to each other, it seems that they are in a personal relationship, one which in time could take on great significance (again, this latter part is of great importance). The interaction could be so meaningful and hope-inducing that it keeps Jones from going insane or perhaps even keeps him from dying or killing himself. Suppose also that, in fact, the tapping is coming from Smith who, many years later, meets up with Jones and they discover what was going on. We submit that this part of their relationship will take on new-found significance in their new relationship, something to look back on and cherish, and a surprisingly good foundation for deepening their relationship now that Jones’s credence has been raised to moral certainty by actually meeting Smith"
• The religious atheist: this ties in with my earlier posts on religious atheism. Take Karen, who identifies herself fully as a Christian. She goes to mass, receives communion, prays, etc, but she seems to be thus made that she cannot believe (in Pascal's terms), e.g., she thinks the problem of evil makes naturalism very likely, or has difficulties believing central Christian doctrines (e.g., the resurrection of Christ). Supposing theism is true, wouldn't we say that Karen has a meaningful relationship with God, even though she does not believe?
• Imaginary friends: There is only limited psychological research on this widespread phenomenon, which occurs quite commonly healthy, normally developing children of preschool age (sometimes continuing into adolescence). Children form what they feel are personal relationships with invisible friends, yet they also seem to be aware that the friends are imaginary. Are these in fact personal relationships? (the difficulty here is of course that there's only one person involved, which is what the theist wouldn't argue) I do not wish to make the analogy between theism and having imaginary friends, only to say that imaginary friends seem to be an example of relationships that do not require belief.