Atheist prayer experiment: Week 4 – What if there’s no answer?

Nothing much happened this week. So, instead, lets talk about divine hiddenness:

It is implied by the rationale for doing the ‘prayer experiment’ that although a ‘positive result’ has value (“Oh, God exists after all!”), a negative result where nothing happens also is a worthwhile result, as it acts as confirmation for one’s atheism. How good is the evidence that no God gets in touch after 40 days for there being no god there?

Divine Hiddenness

God, despite being omnipresent, is pretty hard to find. Given the spread of religious beliefs most people have gotten God fundamentally mistaken; there are also some people, despite being pretty reasonable and open to the evidence, who nonetheless think there is no God out there at all.

These facts have been taken as further evidence that God doesn’t exist. The idea runs something like this. If God exists as advertised by theism, then God has some strong motivations for making his existence unequivocally known: it is much better to go to heaven than go to hell, better to have a religious life than a secular one, and God wishes to be in a relationship with his creation. For all of these goods (and others), knowing that God exists really helps to achieve them. Given God wants there to be good things, then ensuring his existence is beyond doubt seems the sort of thing that should be done.

This is, obviously enough, not what obtains in the real world, as there are all these people who doubt God’s existence. By contrast, if atheism, then the fact there is not evidence that puts God’s existence beyond doubt is exactly what we would expect.

Two examples from these family of ‘divine hiddenness’ arguments. One taking after Schellenburg:

  1. If God exists, inculpable non-belief does not exist.
  2. Inculpable non-belief does exist
  3. God does not exist.

Another slightly different version after Maitzen:

  1. The demographics of theism are very surprising on theism, but not as surprising on atheism.
  2. [Via prime principle of confirmation] The demographics of theism are evidence for atheism over theism

There has been a small hill of literature on this ‘hiddenness problem’. One line of reply is to deny (per popular exegesis of Rom1) that there really is inculpable non-belief: although it looks like there is inculpable non-belief, actually there is some blameworthy motivation against recognizing that God exists. Another is to say that God may have reasons for not revealing himself (at least at a given time), and so we should expect some inculpable non-belief during the course of earthly life. And other things besides.

I’m not going to discuss all of this. Instead, let’s talk about what a ‘negative result’ in the experiment should mean for the atheist undertaking it, and other spectators watching.


On the fact of it, the 70 or so atheists taking part in the experiment are not going to supply particularly powerful evidence for anyone spectating, no matter what happens to them. Other atheists know there are lots of cases of atheists converting to religion day-by-day; believers know there are lots of people who pray and yet lose their faith. So adding a sample of anecdotes of events that are common knowledge is not likely to change anyone’s mind.

It might be different if you know one of these people well. So if you are really convinced that – if only Alice opened her heart to god and started praying – that god would surely allow her to see, then watching Alice sincerely praying away and yet not having god revealed to her might knock your confidence. Similarly, if you thought all atheists were irrational and deluded, interactions with them suggesting otherwise might change your mind. So these ‘exceptions’ might serve to undermine general ‘explaining away’ accounts of people who disagree like ‘they’re all deluded’, ‘they don’t really want to find god’, etc.; but if one does not hold these fully general accounts, then what happens to this small sample is unlikely to surprise one more than what you know to happen in the global population.


Things are different if you are the one experimenting. A negative result in the experiment would help substantiate one of the premises of a hiddenness argument: you know there is inculpable non-belief because you are an inculpable non-believer. You cannot know that for certain – despite your best efforts of insight you rigged the experiment against god and deluded yourself otherwise – but most people will take their efforts to pray earnestly for forty days as discharging their epistemic obligations to ‘seek out God’, and as the actions of an epistemically praiseworthy (leave alone inculpable) non-believer. So they can be extremely confident (and rightly so), that an explaining away account of non-belief per Rom1 is false.

Should it also make one much more confident of atheism in general? It depends. It seems wrong to be wholly certain that God would get in touch after 40 days of prayer if he existed: there may be reasons beyond your ken about why he would not do so, despite that looking like a good thing for him to do (for all concerned). Yet the very fact one takes the prayer experiment implies one things there is at least some possibility of God being the sort of God who would get in touch (this also applies to theists recommending it). So a negative result rules out these sorts of Gods. So if one thinks that – if God exists – he would very likely respond to sincere repeated prayer, a negative result provides one with strong evidence for atheism.

It also might have some practical consequences. Even if one thinks that God is fairly unlikely to respond (but it was worth a shot), a negative response suggests one is either ‘seeking god’ in the wrong way, or that ‘seeking’ is basically futile.[ref]On reflection, this may provide further support for a Schellenburg-style hiddenness argument. If god wants to encourage people to seek him, he at least occasionally needs to reveal himself to the seekers.[/ref] After Mawson’s own analogy: although it might be reasonable to shout in the dark room to see if there really is an old man there, you might stop after the first 40 times: either he was never there, or he is but for whatever reason doesn’t wish to reply to you.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s