In my years at Catholic boarding school, a school chosen by my mother because of its success at persuading young boys to enter the priesthood, I could recite the Credo as readily as the address I scrawled on the envelope of my weekly letter home. “I believe in God, the Father almighty, creator of heaven and earth. I believe in Jesus Christ, his only Son, our Lord. He was conceived by the power of the Holy Spirit and born of the Virgin Mary. He suffered under Pontius Pilate, was crucified, died, and was buried. On the third day he rose again”. It was a formidable collection of beliefs to pack into the ten seconds it took to recite in church every Sunday morning.

My thoughts went back to the Credo when I attended a recent seminar on the subject of belief led by developmental biologist Lewis Wolpert. It was a friendly occasion, hosted by an organisation devoted to scientific debate, and Wolpert was in his usual mercurial form, suggesting a range of reasons, all the way from mental breakdown to the ingestion of hallucinogenic drugs, which might account for people holding beliefs, particularly religious beliefs, which were readily contradicted by scientific evidence.

After his brief, almost cursory, introduction, the largely scientific audience pitched in with their own additions to the list: ignorance of the scientific method, indoctrination, wish-fulfilment (the placebo effect) and a desperate but understandable desire for comfort and consolation. At this point a sort of hiatus occurred. It was as though nearly everyone in the audience was unwilling to ask the one outstanding question. How could it be, even after allowing for all these special factors, that so many people, intelligent, thoughtful people, people in every other respect just like ourselves, were prepared to say that they still believed in such a thing as God, let alone such other yards of attendant nonsense as the virgin birth and the resurrection?

One traditional way out of this dilemma is to suggest that such perverse believers have simply not been properly exposed to the full arguments against their various religious credos. Atheists have been too tentative, too permissive, too polite. We would not happily agree to differ if someone were to assert that the moon was made of green cheese. We would demand evidence. Why therefore should we demur from further debate when confronted by someone who believes that the world was set in motion by an omniscient creator?

But as I discovered some years ago during a university debate, religious believers have well-established ways of dealing with those who persist in invoking science against them. In my opening speech I’d spent some time pointing out the impossibility of reconciling evolutionary theory with the creationist notions which my opponent no doubt espoused. But when my opponent, a relatively distinguished philosopher, rose to speak he simply waved a hand airily in my direction and announced that it was inappropriate to invoke scientific evidence against his beliefs. In a phrase that stuck in my head, he insisted that it was no more possible to prove the existence of God scientifically than it was to prove the world was round musically. I had, he told me with a tolerant smile, simply provided an example of how dogmatic atheists typically confused levels of analysis.

What really seems to be at issue here is the meaning of the word “belief”. In all other contexts than a religious one it seems perfectly acceptable to bring empirical evidence to bear upon what is cited as a belief. Your friend may be irritated when you challenge their belief that Liverpool will win the European Cup with evidence derived from past football results, but they are unlikely to call your intervention inappropriate or out of order.

So what makes religious belief so special, so different? After the Wolpert seminar I searched for some enlightenment in philosophical texts explicitly devoted to the nature of religious belief. How exactly does this belief differ in kind from other beliefs that we hold about the environment, beliefs based on experience, upon empirical evidence?

According to John Hick, one of the world’s leading philosophers of religion, there is no great problem. We have no need to introduce a special meaning of belief to explain religious belief. Like our other rational beliefs, religious belief is based upon experience, upon what we have learned about the world through absorbing information in the normal manner. In his book An Interpretation of Religion he argues that “it is as reasonable for those who experience their lives as being lived in the presence of God, to believe in the reality of God, as for all of us to form beliefs about our environment on the basis of our experience of it.”

Of course we can be deluded about the idea that our lives are being “lived in the presence” of God, but “the general principle on which we operate is that it is rational to regard our apparently perceptual experiences as veridical except when we have reason to doubt their veridicality.” It is, therefore, wholly reasonable for someone who has a “powerful and continuous sense of existing in the presence of God … to be convinced that God exists.”

But even to an amateur philosopher this argument smacks of an illegitimate extension of the word “experience”. It is, for example, not at all clear from Hick’s thesis why we should accord any more special rights to the experience of “living in the presence of God” than we should to the experience of being in touch with the pilots of flying saucers.

I discovered that other prominent philosophers of religion have attempted to solve this problem by suggesting that what is special about religious belief is that it is not so much a belief that God exists, a belief that could properly be susceptible to scientific challenge, but rather a belief in God. In everyday conversation we, and religious believers themselves, often fail to make the distinction. But the difference is significant.

According to James Kellenberger, professor of philosophy at California State University, belief in, unlike belief that, is usually an affective or emotional proposition and typically used in relation to a person. “If a wife believes in her husband she necessarily is not indifferent toward him.” But when we say that we believe in something or more often someone, we are not in the business of making statements about their existence; we are announcing our trust in them. If religious believers took more care with their terminology, if they announced their belief in God rather than their belief that God existed, then they would spend less time at cross purposes with their sceptical scientific critics.

This argument, unlike that advanced by Hick, does at least have the merit of allowing that religious beliefs as popularly asserted (I believe that God exists) are not semantically compatible with beliefs that the sun will rise in the morning or that the boiling point of water at standard pressure is 100 degrees centigrade. But it still seemed to exude an odour of terminological choplogic, an attempt to have it both ways by denying the validity of normal language use.

At this point I felt I’d had my fill of academic philosophy and turned to Michael Frayn’s brilliant but critically underrated book The Human Touch. As soon as I began to read I realised, with a huge sense of relief, that Frayn himself must have been confronted by the very questions that had nagged at me after the Wolpert session. Not only that, but he’d also obviously become equally exasperated by the traditional philosophical attempts to resolve the problem.

Frayn begins his argument by inviting his readers to consider the proposition that “a lot of apparently factual statements … have the same relationship to the world as fictitious ones.” A failure to consider this possibility means that philosophers down the centuries have worried almost exclusively about whether statements “have any meaning at all unless they express a proposition which we know, at any rate in theory, how to establish as true or false.” This leaves fictitious statements “which certainly appear to have some meaning and function … in a philosophical junkyard.”

What sort of belief do I have, for example, in the graveyard meeting in Great Expectations between Pip and Magwitch? One common approach is to suggest that we are prepared to accept the reality of this encounter because, upon opening the novel, we have entered a world of make-believe. We have applied a “willing suspension of disbelief”.

But, says Frayn, is this what is really going on when we read fiction? Do I start off a novel by disbelieving and then knowingly suspend that disbelief? Hardly. What seems to happen is that we provisionally accept an alternative world. It is rather like entering a game. We know the rules and we know what to expect. We believe in a special way.

In this new world fact and fiction can rub shoulders with each other without disconcerting the reader. In the current best-selling novel The Interpretation of Murder, the author, Jed Rubenfield, moves happily between the known facts about Freud’s visit to the United States in 1909 and the entirely fictional account of how Freud became involved in a notorious murder case. As he candidly tells us in a final note: “The Interpretation of Murder is a work of fiction from beginning to end, but much is based on fact.” And yet this juxtaposition of empirical fact and fictitious imagination creates, if well-handled, no difficulties for the reader, no problems of belief or credibility.

So far so good. But Frayn has bigger fish to fry. He suggests that our indifference to the truth-value of what we read in novels is mirrored by the way we behave in everyday life. We don’t go about assigning truth-values to statements in the ways that some philosophers would suppose we do. If I say with true intent, “What a beautiful sunset,” as I gaze out of my window one evening, is this taken by my listeners as an expression of belief? Am I saying it because I believe it? Do I have a set of beliefs in my head, rather like a set of memories, of which this is just one example? Or, don’t I just find the sight of the sun setting over the city to be enjoyable and simply say so?

But compare my statement about sunsets to, say, my declaration that Gordon Brown will be an effective leader of the Labour Party. This looks like a very different kind of belief. On this occasion, my belief suggests doubt. I am not absolutely certain about Brown’s future success but I am nailing my colours to the mast, suggesting that you might like to challenge me. I am issuing an invitation to debate.

And now, finally, we arrive at religious beliefs. When someone says the Credo in church on a Sunday morning, what are we to make of their declaration of belief in such matters as the virgin birth and the resurrection? Are we declaring these beliefs in the same way that we declare a belief in Gordon Brown, suggesting that there might be possible evidence which could contradict or confirm them? Hardly.

This puts Frayn in an excellent position to confront exactly the type of paradox which created the hiatus in the Wolpert seminar. I’ll let him carry the argument: “The unreligious sometimes make the mistake of thinking that the religious can’t understand how oddly the fervour of their believing sits with the factitiousness of their beliefs.” But, says Frayn, this is implausible. “The ranks of the religious plainly include people no less intelligent or perceptive than anyone else. They can see perfectly well how unlike factual statements their assertions are. When they say they believe them, they don’t mean they believe them in the same way as they believe what they say when they tell you the weather is fine, or that there is a train to Birmingham in ten minutes time ... Quite the contrary. It is the pure unevidenced and unevidenceable factitiousness of the beliefs that gives them their force and makes the true believer so eager to believe.” When the religious assert their beliefs they are aiming to secure an effect upon others and upon themselves, to create a moral force field. They are not making statements of knowing. Believers believe. And that is it.

This even opens the way to understanding the extreme, almost absurd nature of some of the beliefs that are being paraded. For the more that such statements are removed from reality, then the less the chance that they can ever be changed by argument or experience. “The stability of religious belief, in which so many believers take pride, is a testimonial to its irrationality, to its insulation from the world.”

Those who have followed the argument to this point may now be feeling an understandable impatience. Do we really need to spend so much time analysing the exact meaning of religious belief if it is fundamentally irrational? When Terry Eagleton complained in his review of The God Delusion that Dawkins lacked any understanding of theology, supporters of Dawkins responded by suggesting that Dawkins’s lack of belief thoroughly excused his ignorance. Would we expect those who disbelieved in fairies to spend time researching fairy-ology?

Perhaps not. But surely if we are to engage with believers in any sort of dialogue we need to have some sense of what they mean when they say that they believe, some sense of how this statement of belief differs from other forms of belief and from knowing. Without some such understanding we will only be furthering a conversation of the deaf. We will constantly be engaged in an enterprise which is philosophically equivalent to an attempt to prove the world is round musically. ■