Praying for outcomes and knowing their value

Some philosophers have been troubled by a general ‘problem of petitionary prayer’: if it's overall best, say, that my friend be healed, God will do it anyway; if it's not best, then He won't; in either case, my asking God to heal my friend won't make a difference to his true prognosis. (Similar worries can be expressed even if the theist denies the assumption that God is required by His nature to do the best.) I take it that, at the general level, the problem of petitionary prayer is solvable by realizing that what might make something best or add to its goodness could in part be affected by whether or not it's been asked for; in a world where petitionary prayer works as the conventional believer supposes it to work, we can get the great good of ‘partnering’ with God. Of course we could have got the good of partnering (and do have this good) in more mundane ways – when, without praying to God about it, we just realize that it would be good to do something and get on and do it. Petitionary prayer gives us, however, in addition to these natural powers for partnership, supernatural powers for partnership, ones wisely mediated by God (in a way they're not mediated in magical worlds). In a world where God answers some petitionary prayers, we gain the great good of being able to partner with God even when confronting issues about which we could do nothing whatsoever through natural means. So, for example, I pray that my friend will be healed of his illness (which, let us suppose, medical science can do nothing to help him with; if there is no miracle, he will die from it); God hears my prayer and miraculously intervenes to heal my friend as a result. This healing wouldn't have happened if I hadn't acted and I couldn't have done anything other than pray to bring it about. So, there's a great good in this world, not just the good for my friend, that he recovers, but the good for me (that my friend is saved by me – I'm a life-saver, with God's help). Further, even supposing God does not intervene and my friend thus dies, it is still the case that I tried to save my friend by praying and to try to save one's friend, even if one fails, is in itself no mean thing.Footnote ¹

So much for prayers of the most usual sort – praying for outcomes that one supposes, indeed very often knows, would be good. There are of course large issues in the background here concerning the issue of whether God must do the best. But we can sidestep them somewhat by observing that all theists agree that God will never do anything bad. He will sometimes allow which of various goods (possibly – on the most exigent view – only when these goods are ceteris paribus joint best) He brings about to be affected by what His creatures pray for (indeed it's their praying for them that, on the most exigent view, breaks the deadlock for the top spot and makes the one better than the other), but He'll never allow a prayer for a bad thing to ‘tip’ His hand, as it were, into bringing that bad thing about. This then brings me towards the sort of case about which I wish to chance my most radical claim. First, though, I wish to consider an intermediary case: praying for something when one is agnostic about its goodness or badness.

Suppose then that I hear of another good friend's falling ill of the same illness as my first friend. This friend, however, is the malign dictator of a despotic regime. I find that I just don't know whether or not his being healed from this illness would be best or even good. But he is my friend; he is suffering; and these two facts alone, when placed in the context of my religious beliefs, mean that it would certainly be natural for me to pray for his recovery, as natural as it was in the case of my morally non-problematic friend. Am I at least permitted to pray that he be healed? It seems to me that I am, if I am careful. To prayers uttered in such circumstances, it is all the more prudent to add the coda, ‘Not my will, but Thy will be done.’ Such a coda has a good precedent and, with it, I am, in effect, saying to God something akin to the following: ‘Allow me to tip the balance in your mind in favour of healing him, God, if indeed there is a suitable balance in your mind (and I don't know whether or not there is). But, of course, if the balance already lies definitively on the side of it being bad for him to be healed (if this isn't one of the issues the value of which can be decisively affected by prayer), don't heal him.’ The ‘Not my will, but Thy will be done’ coda merits some discussion.

There's a sense in which, no matter what one asks for, God's permissive will is going to be done; that's just a consequence of His omnipotence. So, is all one can be doing with this coda noting – perhaps in an approving tone of voice – this fact? Or can one ask for something when one adds it? I think one can do the latter. In some cases, God wills some things to be done only when one or more of His creatures prays to Him that He do them – thus the solution to the ‘general’ problem of petitionary prayer, as discussed above – and this is the sort of case one supposes oneself to be in as one asks for the healing of one's morally non-problematic friend; it is the sort of case which one hopes oneself to be in when one asks for the healing of one's morally problematic friend. But, on reflection, one will know in the first sort of case that one's supposition may be mistaken or in the second that one's hope may be misguided. Is one indifferent over what happens if one is so mistaken or misguided? No, one is not; one wants (and it being God one is sure to get) God's will – as it would have been had one never asked for the first thing – to be done. And one is asking for that with the coda. One is then asking for something that's sure to be granted – He is omnipotent – but one can ask someone to do something that one knows he or she is going to do anyway. My wife tells me, ‘I'm going to open another bottle of wine.’ I reply, ‘Make it a bottle of red if you're indifferent. But it's up to you. Certainly yes, please do open one.’ So it is that I conclude that praying for things that might, as far as one can tell, be good or bad is in principle non-problematic too because, if carefully formulated, in such prayers one's not asking God to bring about something whether or not it would be bad were He to do so; one's asking Him to bring it about if it wouldn't be bad for Him to do so. Things seem to be lining up then such that it will be impossible that one should pray for things that one knows would be bad. But before we get to considering whether or not we can avoid that conclusion, I want to make the point that theists do pray for things that are bad (unwittingly at the time) and are in fact under an obligation to pray for things that are bad (unwittingly at the time), because they're under an obligation to pray and their epistemic resources aren't such as to mean that they'll get their value judgements of the outcomes for which they're praying right all the time. Further, they're knowably in this state and thus those who do reflect enough to know it are obliged knowingly to pray for things some of which will be bad.

As a theist, one finds oneself in a religious equivalent of a variant of the paradox of the preface. Let's suppose that as a philosopher you sit down to write a book. You've just written the word ‘Preface’ when an angel appears and tells you that, if you continue, the work is destined to be one of the greatest works of philosophy ever written. But the angel also tells you that, if you continue, at least one sentence in the main body of the work will assert a falsehood. You know, for the sake of the thought experiment, that the angel speaks the truth. Should you stop writing? Surely not. Even though as a philosopher you loathe falsehood, the price of a falsehood is worth paying for the profound truths that will come with it. We might call the analogous situation if one is a theist ‘the paradox of the prayer list’, the prayer list being a list of everything one will pray for over one's lifetime. Now, in the case of the prayer list, we don't have an angel telling one that something on that list is something that'd be bad, but we have something tolerably equivalent: oneself. If one's relatively young, has (as one supposes it) many years of rich petitionary prayer life ahead of one, and so on, one can know – not know with absolute certainty, perhaps, but know nonetheless – that, at some stage in the future, one will pray for something that would be bad. So, I suggest, theists may, indeed – given the inescapability of our occasionally doing so if we're to continue to have a prayer life at all – should, pray for things that would be bad were they to happen. But of course – like the author not knowing which of the sentences in the book following the preface predicting some falsehoods are in fact the false ones – one (at least in the ordinary run of things) doesn't know as one goes through one's prayer life exactly which of the prayers one is uttering are for things which would be bad were they to come about. If the author came to believe of a particular sentence of his or her book that it was false, he or she would, no doubt, excise it from his or her manuscript before submitting it to the editor, however good he or she supposed that editor would be in editing it out were he or she to leave it in. Similarly, it seems, if a theist came to believe of a particular prayer that it was for an outcome that would be overall bad, he or she would cease to pray for that outcome or at least should cease to pray for it, not continue sending it up to God to let Him ‘edit’ it out. Now then to the final and, I concede at the outset, most difficult case to make consistent with theism: praying for something which at the time one is praying for it one knows would be bad were it to occur. In such a case, one would be acting in the manner of the writer who, knowing of a particular sentence that it was false, nevertheless kept it in the manuscript that he or she submitted to his or her editor. Would this not be the height of irrationality?

Consider the following situation. I have another colleague for whom I have formed something of a dislike, although, from an impersonal point of view, he is as inoffensive as the colleague who is my good friend. No doubt in ways that reflect badly on me, I have even allowed myself to get to the stage at which I sometimes catch myself wishing him quite considerable harm, hoping that he'll fall under the nearest number eleven bus, that sort of thing. He is, in short, my enemy. Now I learn that my enemy has been diagnosed with the same illness as my friends. What would seem to be ideal in these circumstances is for me to pray for my enemy to be healed of his illness. But it is a sad feature of our fallen condition that sometimes we find that we just can't do that which we know would be most spiritually desirable, or at least we can't do it directly. In such circumstances, if we are to make ourselves – by God's grace – able to do it, we'll need to build up to it somehow. A prayer that suggests itself as the obvious ‘second best’ for one to utter in those circumstances is the prayer that God help us become able to do that which would be ‘first best’, able that is to pray for the recovery of our enemies. But uttering even that prayer will not be possible for some people in some circumstances. Let's suppose that I am in this dire position with respect to this particular colleague, my enemy. I don't just hate my enemy; it's worse than that: I don't hate the fact that I hate my enemy; I love that fact. One might suggest that, in such circumstances, I could still pray that I be able to utter the prayer that would be second best. But suppose I love the fact that I love the fact that I hate him; and suppose that it's love all the way down from there: it is, one might say, a love–hate relationship! This is deeply – it couldn't go more deeply – wrong, of course. But it is in the nature of our fallen condition that we can go this deeply wrong. What might then be ‘third best’ for me? Well, it seems to me that someone in that situation might know of his or her own psychology that there are other prayers that God would be more likely to use as a means of assisting him or her in making progress to the point that he or she can do what is spiritually ideal. And perhaps I might know of myself that one of these could be praying for my enemy's death. By way of rendering this plausible, let me step back and describe what I'm going to argue is an analogous case.

Imagine that one day one receives an email from someone who's spotted that one's careful work in the philosophy of religion is being carelessly caricatured and maligned on a popular website by a controversy-seeking blogger and his followers. One visits the site and is, quite naturally, angered greatly by what one reads there. One immediately opens one's email and starts typing up exactly what it is that one would say to this blogger and his mob if one were entirely unconstrained by moral, social, and prudential considerations; one writes what I believe is called a ‘flaming’ email. This is an email which one knows, even as one types it, it would be overall bad for the blogger to receive, but one knows of oneself that one won't be able to turn one's mind to useful things without typing it first. Having spent a few minutes condensing the vitriol, one finds oneself cathartically purged of it, as one supposed one would do even as one typed. One thus deletes the email and is enabled to move on. Now suppose the following alteration to the thought experiment. One knows of oneself that, when one hits ‘send’, one's outgoing emails don't in fact travel directly to their addressees, but are put into a folder which is gone through by one's spouse, who acts as editor-cum-censor; one's spouse lets through to their addressees only those emails that it is at least permissible to let through to their intended recipients; he or she infallibly filters out any that would be bad to let through. In such a case, one wouldn't need to be careful to delete the ‘flaming’ email rather than send it; one could type it up and send it, confident that it wouldn't get through if it would indeed be bad were it to get through, as one knows that it would be. Can I be truly said in such a case to intend to send the ‘flaming’ email to its addressee? It seems to me that I can; all my emails have the feature of going to my spouse first and it would be odd to say that my knowing that of them prevents me ever intending to send any email to its addressee. Of course, I know of this particular email, not just that it will go to my spouse first, but also that he or she will censor it, i.e. that it won't get through to its addressee. But even that knowledge doesn't stop me intending that it do so; I can try to do things which I know I won't succeed in doing; I can ask for things I know I won't get (a couple of examples will follow in a moment); and such is the case here. And writing and sending ‘flaming’ emails is quite plausibly going to be even more cathartic than merely writing and then deleting them. One knows as one hits the send key that the email won't end up getting through to its addressee (though perhaps one hopes that it will; one might hate the addressee and love to hate them, after all), but it may be that it is by putting the issue of whether or not it gets through into the hands of one's spouse that one is best able to let go of it oneself. In the same way, then, I wish to hazard that it could be that praying for a particular thing that one knows would be bad were it to happen should form a part of the prayer life of someone who has – no doubt culpably – fallen into the deepest sort of enmity with his or her fellow man, as, in my latest thought experiment, I have for the person who I have labelled ‘my enemy’.

One may seek to raise doubts about whether the analogous ‘flaming’ prayer can be a genuine petition, rather than something else – merely an articulate expressing of an emotion, for example. But I do not see the grounds on which such a doubt could be well based. They couldn't be simply that one knows of such a prayer that it won't be answered positively. Merely knowing that the person who one's asking to do something won't do it does not disable one from asking for it and reduce one merely to expressing emotion. Perhaps I discover that the college's bursar, having already heard a rumour that I am about to ask him to arrange for the college to reimburse me several hundreds of pounds for a period chandelier that I have purchased for my new rooms in college (to replace the strip light that came as standard), has forearmed himself by obtaining a definitive and irreversible ruling from the college's governing body that the college will not pay for such fripperies. I have already filled out the standard ‘expenses’ form with details of my far-from-standard chandelier and am on the point of delivering it, with its attached receipt and covering letter, to his pigeonhole. I now know that the request will be rejected if I do deliver it, but if I do deliver it, it is still a request. If I deliver it and later ask the bursar if he has received my request, he can hardly deny doing so, saying, ‘Well, I am of course in receipt of the covering letter, expenses claim form, and receipt, but, because I knew that you knew that this “pseudo-request”, as I think of it, was doomed not to generate a reimbursement, I could only interpret these documents (that you may mistakenly think of as a genuine request) as your expressing an emotion.’

Another argument, along similar lines, would suggest that, given that petitionary prayers are those which the person engaging in them must think of as making the outcome prayed for more probable than it would otherwise have been, one cannot pray to God that He bring about some outcome one knows would be bad, because one cannot – in knowing of it that it would be bad – coherently think that one is making it more likely to obtain by praying to God that He bring it about. In response to this, I would say that one may pray for the bad outcome nonetheless, for one can try to do that which one knows one cannot do, employing means one knows will be entirely ineffective in bringing about that at which one aims. In the previous thought experiment, I was trying to get the college to pay for my chandelier by means that I knew would be ineffective. By way of another example, consider the following case. One is suffering from some minor but persistent ailment. Arriving through the post one morning is a free gift, a ‘healing’ crystal; merely having it in one's pocket, one is told by the accompanying leaflet, will heal one of minor ailments. One knows that the crystal will be entirely ineffective. But this, it seems to me, does not make it impossible that one put it in one's pocket and, by putting it in one's pocket, be trying to use it to heal one's ailment. (Remember, we are not talking about the rationality of putting it in one's pocket, but the possibility of putting it in one's pocket and considering oneself by doing so to be trying to heal one's ailment whilst one is also in a state of knowing of it that it will not work to that end.) One might think that the possibility of trying to heal oneself by using a crystal that one knows will not work depends on one's not knowing with certainty that the crystal will be ineffective. If one's really to try to use the crystal to heal oneself, then there must after all be a shadow of doubt in one's mind about its being entirely ineffective. But, even if that's right, knowledge does not require certainty, so this alone won't threaten the conclusion: one might still, knowing (but not being certain) that one will be ineffective in one's intended aim, pray to God that He bring about some outcome that one knows (but is not certain) would be bad. I am not then suggesting with this that such a case reverts to that of the agnostic, as discussed earlier. Far from it: in this case one believes the outcome for which one is praying to be bad and one's belief, I am stipulating, does satisfy the criteria for knowledge. What I am pointing out is that it is not criterial for knowledge that one hold the relevant belief with certainty. Given that, I can concede for the sake of argument that one might need to be uncertain that one will fail in order to try; so be it: one could still know that one will fail. And it is not clear to me that it would be right to say one couldn't try to do something with what would, in ordinary language, count as a high degree of certainty that one would fail to do it. In the two situations most recently imagined, I could try to get the college to reimburse me for my chandelier, whilst being certain that they won't; I could try to heal myself with a crystal whilst being certain that it is entirely inefficacious. So, I conclude that knowing that one will fail to make more probable the outcome one is praying to God for Him to bring about does not disable one from trying to make it more probable by praying to God for Him to bring it about; thus it does not disable one from praying for it. This then is not a reason to think that theists cannot consistently pray for things they know to be bad.

Finally, there's also the point that, given that knowledge doesn't require the impossibility of error, one might reasonably think in the following way about the situation facing one as one deliberates over whether or not to pray for an outcome that one knows would be bad: ‘I know that this would be bad, but it's at least logically (and probably metaphysically) possible it would in fact be good. Therefore I'll pray for it, sticking in the “Not my will, but Thy will be done” coda of course. By so praying if – contrary to what I take myself to know – it is one of those situations where God is willing to allow His hand to be tipped by prayer into bringing the prayed-for outcome about, I'll thus perhaps be able to tip His hand to my liking by so praying.’ This then is not too dissimilar to the justification canvassed earlier when considering the case where one is agnostic about the value of the outcome for which one is praying. In that case, one was praying for it – adding the coda of course – in the hope that it'd turn out to be a situation such as one knew oneself to be in in the first case, a hope which was no doubt easier to sustain given that one didn't know that one wasn't in that sort of situation. In this case, one's prayer may be justified (to at least an extent – more on this in a moment) by the rather desperate hope that one's wrong in one's claim to know that one's not in a situation such as the first case. Of course, given how ‘desperate’, as I'm putting it, that hope is, the costs of uttering this prayer (including of course opportunity costs) would have to be very low (or at least low relative to the gains) for one not to be more reasonable were one to use that time praying for something other than one of the things one knows would be bad. To make the sums add up in such a way that one ends up overall justified in praying for an outcome one knows to be bad, one will need then to buttress this line of justification by pointing to some relatively large gains. However, if we assume to be correct my claim that some people might know that uttering some prayers for things they know to be bad would be, for them, a good way of their moving beyond such prayers, then it follows that, at least sometimes, such gains are there to be had. And, as I hope my analogy of the ‘flaming’ email has shown, that is in itself quite plausible.