Some sophisticated arguments for God have been made in response to the New Atheists. Richard Norman puts the "New Believers" to the test. Since the publication of The God Delusion a common response has been that Richard Dawkins's depiction of religious believers is crude and one-dimensional, and that he misunderstands the subtleties of theology. Terry Eagleton, for example, argues that Dawkins's picture of belief is a "vulgar caricature", while Karen Armstrong suggests that Dawkins misapplies logic in the realm of myth, where it has no business, and that he is ill-equipped and unwilling to consider the essential poetry of belief. They, and others, argue that believers are not the literalist bunch of irrationalists lampooned by card-carrying rationalists, but a diverse group of questers after truth. For them God's house has many rooms; it is a well-stocked library with spaces for deep thought and rational argument, for myths and stories and poetic metaphors, which Dawkins, and those who agree with him, have failed to read.

As I have argued before, Dawkins does over-simplify. Although he knows perfectly well that most Christians are not creationists, he sometimes writes as though they were, and implies that all religious belief is just obviously refuted by science and Darwin. He is inclined to treat all versions of religion as equally irrational. Suppose, then, that we take Eagleton, Armstrong and the other "New Believers" at their word, and that we try to do justice to the nuanced diversity of the views of the religious. Does that get them off the hook?

Let's first consider the relationship between religion and science. Some argue, as Eagleton did to Laurie Taylor in New Humanist 4/2009, that religion and science are completely different kinds of things: "Christianity", Eagleton avers, "was never meant to be an explanation of anything in the first place." This will come as news to millions of believers, and would have amazed all those for whom, down the centuries, the appeal of religion has been that it offers answers to questions they can't otherwise answer – "What is the origin of the universe?", "Why are we here?"

It would also surprise those of the New Believers whom we might call the "rationalists". Keith Ward, for example, the former Regius Professor of Divinity at Oxford, in his book Why There Almost Certainly Is a God (subtitled Doubting Dawkins), argues that there is a sound rational case for believing in the existence of a God, because "the God hypothesis" provides the best explanation of why the universe is as it is. He doesn't, of course, think that "the God hypothesis" is a scientific explanation. What he and others claim is that scientific explanations are incomplete, and therefore need to be supplemented by a different kind of explanation.

Ward, in company with other rationalists such as Richard Swinburne and John Polkinghorne, defends the "fine-tuning argument". Given the fundamental scientific laws and the basic physical constants, they say, we can explain scientifically how the "Big Bang" gave rise to our universe, and how in due course this led to the emergence of life and the evolution of species on our own planet. But this leaves unexplained why the scientific laws and physical constants are as they are – for if they had been even slightly different, the Big Bang would not have produced our universe or any ordered universe at all. The best explanation of why the initial conditions were just right for producing our universe and the creation of life is, they say, not a further scientific explanation but a "personal" explanation, one which invokes the purposive intentions of a creative intelligence.

Ward and Swinburne and others say that this is a good explanation because it is the simplest explanation. Dawkins rightly replies that this is to misunderstand the requirement of simplicity. Granted, the hypothesis of a creative intelligence is simply stated. But to claim that there must exist something or someone with a mind capable of understanding all the consequences of all the scientific laws and the initial conditions is to posit the existence of a being so vastly complex that it stands much more in need of explanation than what it is supposed to explain.

We cannot just assume that the only good explanations are scientific explanations. We need to take seriously the claim that scientific explanations are incomplete, and need to be supplemented by a different kind of explanation. But what we can properly insist is that any proposed alternative kind of explanation must still meet the same standards for what counts as a good explanation. In particular, a good explanation can't be one which makes things even more inexplicable.

So where does this leave the debate between Dawkins and his rationalist opponents? Dawkins says at one point: "Christianity, just as much as Islam, teaches children that unquestioned faith is a virtue. You don't have to make the case for what you believe." That's much too sweeping. By the very act of producing counter-arguments, Dawkins has to acknowledge that some Christians, at any rate, do make a case for what they believe. It's just that their case isn't good enough.

So much for the rationalists. For Terry Eagleton this is all irrelevant; religion, or at any rate Christianity, is not a belief in the existence of some kind of supernatural entity, it's not an explanatory hypothesis and so it doesn't need to be either rationally defended or accepted on faith. I think he's wrong about what Christian belief has traditionally meant for its adherents, but still, this kind of sideways move needs to be taken seriously. It's an attractive move for those who want to maintain that religion is not threatened by scientific progress. Karen Armstrong, in her recent book The Case for God, goes even further. Like Eagleton she's a lapsed Catholic who still feels protective of her ex. For her, religion as traditionally understood and practised was not a matter of belief at all – or at any rate not belief in what she thinks is the purely modern sense of intellectual assent to factual statements.

If religion is not about belief, what is it? It's worth looking briefly at three suggestions for what might be central to a "belief-lite" version of religion. The first is that religion is primarily a matter of rituals and practices, rather than beliefs. Armstrong is keen on this idea – she thinks that religion has its origin in rituals. John Cornwell, in his book Darwin's Angel, chides Dawkins in similar terms:

"You think religion is 'a persistent false belief held in the face of strong contradictory evidence'. I am sure you realise that this hardly exhausts the meaning for most people ... Religion, of course, is only partly involved in actual beliefs or doctrines ... Religion's activities, its rituals, its mythologies, hymns, meditations, prayers, chants, poetry, images, parables, legends, taboos, and sacramentals (by which I mean holy objects, such as candles, incense, oils, vestments, holy water) are principally symbolic."

Of course it's true that a religious community is not a debating society, and a religious service, with its prayers and hymns and rituals, is not an academic seminar whose business is to assess and defend theories. But the key phrase in the passage from Cornwell is "only partly". The asserting of beliefs may not be the main preoccupation of religious activities, but it is still essential. Without the beliefs, the practices make no sense. Prayer is meaningless without a belief of some kind, however vague, that there is someone, a person, who is being addressed. Hymns of praise and adoration are meaningless without some kind of belief in a deity who is worthy of adoration. And there is accordingly no evading the question of whether these beliefs are true. When I listen to, and am emotionally overwhelmed by, a performance of Bach's B Minor Mass, I am presumably experiencing something different from a devout Catholic attending Mass. And the difference must surely lie in the beliefs.

Here's another suggestion which Armstrong and others are keen on – that religious language is not intended to express straightforward factual beliefs because it is metaphorical. Again there's a lot to concede here. Talk of "God the Father" is not about a biological relationship, and as Eagleton says, intelligent believers don't think that God "created the world rather as a carpenter might fashion a stool". We should accept that, unlike the crude views of the fundamentalists, intelligent religious believers have always seen themselves as groping for language to express the inexpressible.

Still, the language of religious belief cannot be all metaphor. We must be able to say at least something in non-metaphorical language to identify what the metaphors are about. Consider the opening of the famous poem On the Nature of the Universe by the Roman poet and philosopher Lucretius, writing in the first century BCE: "Mother of Aeneas and his race, delight of men and gods, life-giving Venus, it is your doing that under the wheeling constellations of the sky all nature teems with life, both the sea that buoys up our ships and the earth that yields our food. Through you all living creatures are conceived and come forth to look upon the sunlight."

We know, from everything else in the poem, that Lucretius is an atheist and doesn't believe that the goddess Venus really exists. She's a metaphor for the creative power of sexual desire in the natural world. Now compare this passage from Psalm 104: "O Lord my God, thou art very great... Thou waterest the mountains; the earth is satisfied with the fruit of thy work. Thou dost cause the grass to grow for the cattle, and plants for man to cultivate, that he may bring forth food from the earth, and wine to gladden the heart of man, oil to make his face shine, and bread to strengthen man's heart."

Of course there's metaphor here: God is not literally a celestial gardener with a giant watering-can. But it can't all be metaphor. The psalmist is not just talking about the creative power of nature. What separates him from Lucretius? It can only be the non-metaphorical belief that there really is a divine being, possessing powers which are in some way comparable to human powers of intention and understanding, however much they surpass them, who is the ultimate explanation of natural processes.

Here's a third suggestion – that what the traditional religions offer are stories, myths, to be understood not as literal factual reports but as imaginative fictions. This is what Armstrong is especially keen on, and she has good things to say about it. I agree with her that stories play a vital part in helping us to make sense of our lives, and provide a kind of understanding which is quite different from scientific explanation. But why confine ourselves to religious stories, myths? There are wonderful stories in the Bible – the Garden of Eden, Noah and the flood, Joseph and his brothers, and so on – and they can give us valuable insights, but can't they just take their place alongside all the other stories, from the myths of ancient Greece and Rome to modern novels?

In her earlier book A Short History of Myth Armstrong accepted that novels can play the role that myths have traditionally played. "A novel, like a myth, teaches us to see the world differently; it shows us how to look into our hearts and to see our world from a perspective that goes beyond our own self-interest." But in The Case for God she reverts to the idea that religious myths have a special status. And of course every particular religion privileges not just myths in general but its own set of myths. Being a Christian, if it's a matter of stories, means guiding your life by the Christian myths. Traditionally these myths have been backed by factual beliefs which they are supposed to illustrate, beliefs about a divine being who, if he didn't create the world in six days, is at any rate the creative power behind the natural world, and historical beliefs about the life and death of Jesus of Nazareth. But if all that we've got are myths, then why privilege the stories of a particular religion?

Because they are our stories, it may be replied – and this brings me to one last way of downplaying the role of beliefs in religion. Nicholas Lash, in his essay "Where Does The God Delusion Come From?", accuses Dawkins of "taking it for granted that to 'believe in God' is to be of the opinion that God exists." That's a travesty of religious commitment, he says. Rather, "To be a Jew, or a Christian, or a Muslim, is to be a member of a particular people: a people whose identity is specified by particular habits of memory and ritual, of understanding and relationship and hope."

Factually, this may be correct. Most people give their allegiance to a particular religion not because they judge that its doctrines are uniquely true, but because it is the religion in which they find their cultural roots. It defines their identity. And it does so through a particular set of practices and rituals, and a particular set of metaphors, and a particular set of stories. And it is this above all, I think, that explains why, for all the intellectual implausibility of the traditional doctrines, intelligent and sensitive people continue to identify themselves as Christians or Muslims or whatever.

I offer three comments on this view of religion as a matter of identity rather than belief. First, this version of religion may be relatively benign. Someone who sees their religion as primarily a matter of powerful metaphors and myths is far removed from the idiocies of fundamentalism and creationism. And because they are not committed to a literal interpretation of their scriptures, they are less likely to find in them a divine injunction to persecute gays or oppress women or blow up buildings or kill doctors who perform abortions (though there may still be institutional pressures on them to do at least some of those things).

A second, more negative comment: is the religion of identity really what we need right now? In a world of sometimes murderous conflicts between people of different creeds and cultures and ethnic groups, shouldn't we be looking for the common ground on which we can come together, rather than the things which cement those separate identities?

My final comment is, in a way, the opposite. A religion built around metaphors and stories, rather than doctrines, seems to me to be inherently unstable. If talk of divine creation is just a metaphor for the awe-inspiring beauty and complexity of the natural world, it can hold that meaning for anyone. The stories are a common heritage. We all share in the riches of a culture which includes not only the Bible but the great literary and artistic treatments of the Bible stories – Paradise Lost and Haydn's Creation, the glories of Renaissance art, the life and death of Jesus in Handel's Messiah and the Bach Passions – and, increasingly, the images and symbols and art of other religions.

"We" here includes humanists. So what is it that sets the religious apart? Isn't an identity based on metaphors and stories always going to be fragile and porous? I cannot see how, in the end, a distinctive religious identity can be possible unless it is based on the acceptance of at least some non-metaphorical factual beliefs – beliefs about the existence of a personal deity and about how his intentions and purposes explain our world. Those beliefs do, inescapably, need to be rationally defended. And they can't be. On that point, certainly, Dawkins is right.

A longer version of this essay appears in the collection Religion, to be published by Cambridge University Press in 2010.

Published 2009-11-17

First published in New Humanist 6/2009

Contributed by New Humanist
© Richard Norman / New Humanist
© Eurozine