In this paper, I aim to discuss not the issue of whether or not we do in fact have reasons to suppose that there is or that there is not a God, but rather an issue which looks at first glance like it might have a certain methodological priority, the issue of what is the right ‘ethics of belief’ for belief in God: should one believe in God only if one has positive reasons in favour of doing so or is it permissible to believe in Him without such reasons? In this paper, I want to see what can be done to decide which of the various options for an ethics of belief in God is right prior to deciding whether or not God exists. Let's start then by getting these options out onto the table. One of them is as follows:

1. 1) We should believe in God only if we have positive reasons to do so.

Many atheists hold that this is the right ethics of belief for belief in God and, thinking that there are no such reasons, they thus think of themselves as doing as they ought when they remain atheists. But many theists hold that this is the right ethics of belief as well; the difference between many theists and atheists then is just that the theist thinks that there are in fact good reasons for believing that there is a God and the atheist thinks that there are not. So it is that this shared ethics of belief lies uncommented on in the background of many public debates between prominent unbelievers and believers – you can assume that it does whenever the debate focuses on advancing and attacking arguments for the existence of God. But option 1 is not the only option.

There are some who believe in God even though they do not take themselves to have any positive reasons for doing so and who do not consider themselves in any way intellectually irresponsible in so believing. Deciding whether or not there's a God, such people say, is a ‘leap of faith’, rather than a conclusion of reason. The most moderate variant of this ethics of belief we might summarize as follows:

1. 2) We are permitted to believe in God even if we don't have any positive reasons for doing so, as long as we don't have any positive reasons against.

Obviously, on the first view, in order appropriately to believe in God, one has to have both positive reasons for doing so and solutions to arguments which would purport to overwhelm these reasons with reasons against (e.g. the Problem of Evil). But on this second view, it is alright to have no positive reasons for believing in God, just so long as one still has those solutions to things like the Problem of Evil. On this second view, the hurdle is lower. As you might expect then, many atheists object to this ethics of belief. Indeed many seem to find it infuriating and quickly lose patience with people who seem to be espousing it. But if people who espouse 2 set your teeth on edge, the next group we'll consider will make you wish you'd had them taken out.

There are people who hold an even more extreme ethics of belief than 2, one which suggests that we are justified in believing ‘by faith’ things which are actually contrary to reason. We might summarize that view as follows:

1. 3) We should – or at least are permitted to – believe in God even though, or maybe even because, it is contrary to reason.

On this view, at its most extreme, belief in God becomes appropriate proportionate to the extent that one has reasons against it. No need now then even for a solution to the Problem of Evil, indeed perhaps better not to have one for then one can revel all the more in the glorious absurdity of one's faith. On this understanding, the hurdle isn't just low, it's below ground. Atheists who wish to attack this as the right ethics of belief for belief in God find themselves making common cause with many theists, viz. all those who hold either of the first two ethics of belief.

The three options for one's ethics of belief that I've enumerated above seem to be mutually exclusive and exhaustive, that is to say that at most one of them can be right and at least one of them must be right. Let's look at these options in a little bit more detail to see whether we can vindicate the hope that we can decide which one is right prior to deciding whether or not there's a God. The first option was:

1. 1) We should believe in God only if we have positive reasons to do so.

How might one defend this? One might base a defence of it on something like W. K. Clifford's principle, ‘it is wrong always, everywhere and for anyone, to believe anything upon insufficient evidence’ (from W. K. Clifford, ‘The Ethics of Belief’). But that principle would cause too much collateral damage: it would undermine the propriety of our believing things we all think it is permissible to believe and yet for which we have no evidence at all. For example:

It seems that one can generate sceptical scenarios (we're all disembodied spirits being fooled by a demon; we're all brains in vats; et cetera) that are indistinguishable from the points of view of the people in them from being in the world we take ourselves to be in. In that the differences between these scenarios and what we take to be reality are then ‘evidence transcendent’, so we cannot have any evidence – let alone sufficient evidence – against these scenarios being false. Yet we all do, nevertheless, think them false and do not think ourselves irresponsible in so thinking. Clifford's principle is too strong.

To avoid this sort of collateral damage, yet successfully hit belief in God, one could seek to narrow the focus of the principle in some non-question-begging way. But, it seems to me, this can't be done. So, for example, one might say that the principle applies only to topics on which there's widespread disagreement amongst sincere, intelligent and well-informed (at least about areas other than the one in question) people. This then would allow back in as entirely proper our commonsense beliefs that there's an external world and so on, but still keep out belief in God.

However, even with this narrowing, the principle still causes too much collateral damage; all our fundamental moral beliefs get hit, one example being our belief in the principle itself. The modified Cliffordian principle is itself something on which there's widespread disagreement amongst sincere, intelligent et al. people. So, if it's true, it cannot be properly believed to be so unless one has sufficient evidence in its favour. It seems one can't provide that. So it looks as if option 1 is one we can't have reasons to believe is true, or at least reasons to believe is true prior to collecting reasons to believe that there is or that there's not a God. It might of course be true anyway. (Something's being something which, if true, is something which one cannot find reasons to suppose is true is not that thing's being false.) Let's leave it on the table then.

What about 2? To remind ourselves of what 2 was:

1. 2) We are permitted to believe in God even if we don't have any positive reasons for doing so, as long as we don't have any positive reasons against.

How might one defend this? Again, the issue in defending it is avoiding collateral damage, i.e. we don't want to end up ‘defending’ it whilst inadvertently defending the obviously indefensible (and thus not really defending it at all). But, this time, the danger is not that we might keep out too much, but that we might let in too much.

So – to use an example of Russell's – we do not think that we are permitted to believe that there's a teapot orbiting the Sun (even though – I take it – we don't have any positive reasons against the hypothesis that there is for careful enough statements of the hypothesis). Thus we'll need to come up with something special about God viz. a viz. teapots and the like, something special that isn't question-begging and which allows us to believe in God but disallows us to from believing in Russell's teapot.

William James, responding directly to Clifford, seems to me to have done one of the best jobs here (see his ‘The Will to Believe’), but – to cut a long story short – it doesn't seem to me to be conclusive. That is to say that even whilst it is no doubt true that there are some things that we are permitted to believe (even should believe) when we don't have any positive reasons for doing so – we might go back to our examples of the beliefs that the future will resemble the past; that we weren't all created five minutes ago with a host of false memories; and that there is an external world and it is more or less as our senses present it to us – the belief that God exists does not seem to me to be able to be categorized amongst these on grounds independent of God's existence.

To expand a little on that last point: it seems to me necessary that there are some things which we are permitted to believe (even should believe) even though we don't have any positive reasons for doing so. If that weren't the case, then, for every one of our beliefs, we'd need another belief (to be the required reason backing it up), but that would generate an infinite regress and we cannot have an infinite number of beliefs. If then believing some things without any positive reasons in their favour is intellectually inescapable for finite minds such as ours, it is intellectually responsible; we are at least permitted to do it, probably should do it. The only real question can be over what beliefs fall into this category of beliefs, which we may call ‘properly basic’ beliefs.

Whilst there's unlikely to be much controversy over putting things like ‘there's an external world’ into the category of properly basic beliefs, there is likely to be controversy over putting things like ‘God exists’ into it. It seems that if there is an external world, then belief that there is such a world is likely to be properly basic and, given that we all do believe that there is an external world, so we can all agree that believing that there is without any reasons is at the least permissible. Similarly, it seems that if there is a God, then belief that there is might well be properly basic too (indeed probably is properly basic for at least some people – God himself, for example), but, given that we do not all believe that there is a God, so we cannot expect that we'll all agree that believing that there is without any reasons is at the least permissible.

It seems to me that the contemporary philosopher of religion Alvin Plantinga is right when he says – roughly – that if God exists, then we are permitted to believe in Him even if we don't have any positive reasons for doing so; and, if God doesn't exist, then we're not. It looks to me then as if 2 is something we can't have reasons to believe is true either, or at least reasons to believe is true prior to deciding whether or not there is a God. As with 1, it might be true none the less. It too needs to stay on the table. What about 3?

3 was as follows:

1. 3) We should – or at least are permitted to – believe in God even though, or maybe even because, it is contrary to reason.

Hurrah. At last, we've found a view that we have reason to reject: it's impossible to articulate the view without giving oneself a reason to reject it. (This explains why it's hard to find anyone who uncontroversially espouses the view: Tertullian? Kierkegaard?)

3 is saying that we should (i.e. have overall reason to) – or at least are permitted to (i.e. don't have overall reason not to) – do something even though, or maybe even because, it is contrary to reason, i.e. we have overall reason not to do it. But the claim that we have greater or equal reason to do something which we at the same time have greater reason not to do is a straightforward contradiction. It's a logical impossibility that this obtain; thus it is that nobody can non-confusedly hold that it does obtain, which is why Tertullian and Kierkegaard are probably most charitably interpreted as being ironical when they appear to state this as their view.

So we may conclude that 3 is out. The right ethics of belief for belief in God is either 1 or 2. And we may conclude that our initial hope that we could decide which of these is right methodologically prior to deciding whether or not there's a God is misguided; we can't. 1 is right if there is no God, but 2 might be right – indeed almost certainly is right for some at least – if there is a God. What to do next then? It looks as if it'll need to be Natural Theology (that is to say we'll need to study those arguments which purport to give us reason to think God exists) and Natural ‘Atheology’ if you will (that is to say we'll need to study those arguments which purport to give us reason to think God doesn't exist). But what might otherwise have seemed a strange result may now be anticipated as at least possible.

If we've followed my line of argument here, we'll now be open to the possibility that if our studies of Natural Theology and ‘Atheology’ were to end up giving us reasons to believe that there is a God, then they might end up giving us reasons to believe that these reasons for belief in God weren't after all necessary. That fact – the fact that we might, by finding reasons to believe that there is a God, be finding reasons to suppose that we don't need these reasons – might explain why some of those who do believe in God are so indifferent to the fact that they can give no reasons for their belief. And realizing that might make those of us who are atheists and subscribe to option 1 a little bit more patient with those of us who are theists and subscribe to option 2.