Personal update: I have been finding praying pretty difficult in the last week or so. It has been hard keep my mind on prayer rather than wandering elsewhere after the initial mental recitation. Similarly, I have generally failed to keep the regimented pattern I would like – not every evening, but scattered throughout the day.

I have noticed nothing. Life continues as before, and nothing in particular has suggested that God is smiling on me or getting in touch – if anything, I’ve been mildly more worldy and stressed than the usual. Time will tell.

My college chaplain has the misfortune of knowing me, and she recommended I give some other ‘ways’ of praying a go: lectio divinia, and turning up to evening prayer. Being kergmatically ‘up for anything’ I have tried the former this week, and will try the latter soon.

To pass the time, let’s talk about the philosophical rationale behind this ‘Atheist Prayer Experiment’.

Praying to stop being an Atheist?

Mawson is a philosopher of religion based in Oxford. His paper “Praying to stop being an atheist” is one of the motivations behind this experiment. Most of the test subjects aren’t particularly keen on the line of argument, even if they are willing to give it a go regardless. I think Mawson’s paper is basically right (modulo some minor caveats), so I’m going to do my best to exegete and defend him.

The argument runs something like this. If you are an atheist who considers the probability of God existing as low (but not negligible), and you think that prayer could be a truth directed practice (that if you prayed and heard God answer, this would at least provide evidence for God existing, and not be an obvious ‘false positive’), and the costs of praying are low, then praying to God, even if an atheist, is a good thing to do.

Let’s unpack each of these in turn.

God’s probability and the expected value of information

There are lots of ‘prayable’ entities we can dream up: fairies at the bottom of the garden, the Invisible Pink Unicorn, the Flying Spagetti Monster, etc. Praying to all of them is infeasible. So why should we privilege God above all of these other alternatives?

Two-ish reasons: the first is that, in fact, the IPU, FSM etc. have a negligible probability, whilst God has a non-neglible probability. Why say that? Because the FSM/IPU/etc. are parodies, and parodies seldom happen to be true. Of course, the point of these parodies was to say that God is ‘just as (un)likely’ as these other things, but that is false: the fact God is sincerely believed by a lot of people, including many who appear reasonable and bright provides at least some evidence in favour – evidence lacking for the FSM/IPU. I’d be reluctant to bet at 100/1 odds against God, but happy to bet at 10000000/1 odds against the IPU. Pace some really good knock-down argument for atheism, the probability of God should not be taken as non-neglible.

The second reason is noted by Mawson: God stands apart from the other ‘prayables’ in that his existence is a matter of considerable importance. Here’s a few reasons why it is important to ‘get it right’ about God:

1. If God exists, believing the right things about God may entail eternal consequences like getting into heaven and avoiding hell.
2. If God exists, forms of religious praxis or experience (again, depending on religion) will greatly benefit you – spiritual reassurance, prayer, divine love, or whatever else. These benefits may not be accessible in the course of a ‘secular’ life, even one that flourishes in all other respects.
3. If God exists, things like prayer or scripture may provide guidance on important moral issues, which may help one in cases where one is unsure of what the right thing to do is.

So, from our atheist point of view, God seems to be the only intersection between ‘important if in fact existing’, ‘non negligible probability of existing’, and ‘might answer prayer’.

Mawson urges us to think prayer is a very high value-of-information activity. If one receives an answer suggesting God exists, then one should reconsider how to live and believe; if one receives no answer, still more reason to be confident in one’s atheism. Either way,the reward of getting more salient evidence on such a vital question is worth the pretty small costs of prayer.

Prayer and the risk of self-delusion

Prayer might be a bit more costly than that. One might worry that praying for an answer might lead one to see one that isn’t really there: one might simply delude oneself that coincidence was an instance of divine intervention, or that an otherwise mundane ‘funny feeling’ is reified into mystical experience. Although one may argue the costs of wrongly believing God is there are not as great as wrongly believing God isn’t there (compare Pascal’s Wager), there still are some costs to a ‘false positive’: it might lead you to wrong beliefs (ethical or factual) about the world, it may lead one to devalue what is truly valuable in life in preference for chasing after a false numinosity, etc. So prayer might be playing with (self delusory) fire.

Mawson foresees this worry. He accepts there is inevitably some risk of ‘false positives’: no matter how well we think we know ourselves, there is some chance of being mistaken and deluding ourselves. Yet it is bad epistemic practice to avoid all epistemic methods that have any risk of false positives: rather, we should avoid methods that have substantial risks of false positives. For most atheists, Mawson urges, they can be confident that they are not at substantial risk of false positives:

[W]e must in each of our own cases individually assess how plausible is the supposed psychological mechanism whereby we risk being rendered dotty, generating false positives for ourselves; and I hazard that, for most agnostics and atheists reading this, it will not be very plausible at all. Most agnostics and atheists are not that suggestible; they will (rightly) believe probably false of themselves claims such as the one that were they to engage in this practice of prayer, even over a relatively elongated period, they would run a significant risk of coming to the belief that there is a God even if there’s not through, for example, ‘projecting’ out into a hallucination a sublimated father-figure. Of some weak-minded people, they will believe, such claims may be true, but they will believe that they are not true of them. And, I suggest, most of them are right. (pp 179)

I don’t think this worry can be dismissed quite like this. Although some atheists might take believers to form their beliefs via projection or whatever else, other sorts of atheists might be more friendly (epistemic sense) about it. They may hold that what separates them from believers is not flagrant irrationality or self-delusion, but more minor epistemic misdemeanors: over-interpreting certain personal events, being too trusting of their friends testimony, being a little too pliable to community pressure and norms, and so on.

The problem is the friendlier the atheist, and (to use Mawson’s phrase) the less ‘dotty’ they think one needs to be to believe one is communicating with God even though he doesn’t exist, the more serious the concern about ‘going dotty’ and getting false positives become. Ironically, unfriendly atheists (who take religious conviction to be pretty irrational) should rush in to this experiment, whilst friendly atheists (who may think religion is mistaken, but well within the envelope of mistakes very sensible people can make) should fear to tread.

But even for those friendly atheists, they should still step forward carefully. Providing they have some idea of what sort of errors in reasoning foster mistaken religious belief, they should prosecute their experiment to avoid them: if they think much religious conviction is forged by the psychology of communal worship, they should avoiding ‘experimenting’ with prayer in these sorts of contexts. So long as they can satisfy themselves that a particular ‘method’ avoids (or avoids-enough-so-risk-reward-ratio-is-good) the non-truth-directed drivers, then it is still in their interests to pray.

There may be some atheists who ‘can’t put their finger on’ what is exactly is going wrong with believers, such that they cannot be convinced that any prayer experiment has a low enough risk of false positives to be ‘worth it’. However, I take this to be a fairly small minority – besides, if you think a group is reasonable, but somehow getting things wrong (but you don’t know why), this might give some reason to think you, not they, are the ones mistaken about things.