Folk beliefs unite humanity, while religious beliefs divide us. A. C. Grayling asks why we believe and pits intellectual honesty against the draw of the deity.

Everything anyone does is motivated by beliefs, conscious or otherwise. Because of the complexity of the world and the human mind, beliefs are many and various, ranging from the stock notions of "folk physics" and "folk psychology", which respectively concern how medium-sized physical objects behave and what other people's smiles and tears generally mean, to sophisticated (and sophistical) systems of political and religious beliefs that motivate not only individuals but whole nations.

Beliefs themselves can be placed in two broad categories. On the one hand, there are those that are held on the basis of an original acquisition in rational reflection upon observation and experience, and are always subject to test by further observation and experience, to be rejected if they fail.

And then, on the other, there are those grounded in non-rational sources, such as superstition, tradition, authority, emotional need and the like, which individually or collectively constitute one main source of religious belief. This second category might also include those beliefs that satisfy the first condition of the first category (early man might, in his state of ignorance, have reasoned that thunder was the sound of large, powerful and invisible versions of himself walking about in the sky), but not the second. Herein lies religion, early man's science and technology, which has endured despite the evidence of infinitely repeated disconfirmations of its claims and contradictions within them.

É«ºÐÖ±²¥

Most day-to-day human behaviour is premised on the "folk" beliefs that constitute our basic understanding of the physical and social worlds. These are widely shared across cultures and time for the good reason that they relate to the natural cognitive endowment of human beings, and the fact that we are social animals. A mixture of hard-wiring, empirical encounter with the world and social sharing of the fruits of experience via language, explains most of the content of these universally held beliefs. They tell us that fire is too hot to touch, that water slakes thirst, that people laugh when happy, that dropping a large stone on someone's foot will cause a physical injury and annoy the person psychologically, and so on. These beliefs operate in the daily lives of present-day New Yorkers and Kalahari Bushmen alike, and between them both and ancient Abyssinians.

What divides groups of people from one another, by contrast, are the conscious political and religious beliefs that their leaders or traditions supply, in most instances, by imposing them - and in the case of religion, by inculcating them in intellectually defenceless children. To create and maintain distinct groups, concepts of identity are required. These in turn require special sets of beliefs that explain the difference between one's own group and others, and (typically) the superiority of one's own group over others. Historically, the most powerful source of these adventitious identities has been religion.

É«ºÐÖ±²¥

There have of course been other identity-conferring devices: citizenship, skin colour, language and gender among them. In the modern West, combinations of these factors have become increasingly significant as the importance of religion as a differentiator and separator has diminished.

Nevertheless, religious beliefs have remained a potent identity marker.

Protestant, Catholic and Jew have lived largely segregated lives even within today's Europe, and Christians have set about murdering one another and Jews in highly organised ways with alarming frequency. Likewise, Christians and Muslims have often fought, the former putatively to regain their faith's holy sites, the latter to conquer and convert the former.

There are two different tasks at issue in discussing the motivating role of beliefs, one descriptive, one evaluatory. The aim of the first is to tell us how belief translates into action or further belief, and accordingly provides ways of inferring people's beliefs from the way they act. Although acting and dissimulation can mislead us in judging what others believe, in typical cases the mechanisms are straightforward. If a hunter tracking a deer in a forest tiptoes up to a bush and fires his gun into it, then - given his aims and intentions, and the overall point of his activity - the probability is that he believes the deer to be hiding there. If he said that he had thus tiptoed and fired but neither believed the deer was hiding there nor had an alternative explanation for his act, we would think him very peculiar.

Descriptive theories of belief tell us much about the nature of major emotions such as anger, pride, love, hate and guilt by identifying the kinds of beliefs that operate in producing them and the behaviour they typically prompt. Shame, for example, is felt by those who feel they have failed according to some standard, or have done wrong according to some code, which involves sets of beliefs about the validity of the standard or code in question, and their own standing in relation to it. Shame also involves what the individual believes others are likely to think about them in the light of that failure, and what they might do about it.

Much of what can be said in the way of descriptive theories of belief and action is common sense, because most normally competent human beings are adept at explaining their own and others' behaviour in the light of their shared beliefs. But there is plenty of room for inquiry into more complex states, such as unconscious belief and motivation, especially when it is pathological in some way, prompting people to maladaptive behaviour of all kinds from the eccentric to the dangerous. Some theorists - Freud is an example - think that we are systematically wrong in what we assume about our beliefs and actions, which have to be understood (with psychoanalytic help) by reference to unobvious wellsprings.

Equally interesting, but for our present purposes more important, is the second task mentioned: that of evaluating the motivating role of belief.

Here the central question concerns what has been called "the ethics of rationality", which among other things is about one's responsibilities in holding, acquiring, reasoning about and acting upon beliefs. As the word "ethics" implies, the task is one of identifying what one ought to do if one is to be - so to speak - a good believer and a rational actor.

É«ºÐÖ±²¥

The primary quality required by anyone ambitious to be a good believer is intellectual honesty. This is because, without it, the primary duty of a good believer cannot be fulfilled - this being to subject to scrutiny every belief already held and any in the process of being newly acquired, with the commitment to reject it if it does not meet demanding standards.

Intellectual honesty is a commodity in exceedingly short supply. Too many people wilfully ignore inconsistencies between various of their beliefs, and evidence that controverts them. Too many are lazy about thinking things through, testing what others say, checking facts and challenging themselves over the reasons they have for wishing to believe one thing rather than another. Too many deceive themselves, or substitute wishful thinking for rational evaluation, or barricade themselves behind spurious arguments or the authority of convenient others.

Very few religious people who made a sterling effort to eschew these common vices would be able to sustain a commitment to belief in any of the world's faiths, major or minor. Perhaps that is why very few make that sterling effort in the first place. It is certainly why the major religions ensure that they sow the seeds of credence early, in the minds of children unable to evaluate what they are told to believe, in the expectation that this will, in at least many of them, make intellectual honesty usefully impossible.

The demanding standards that intellectual honesty imposes are in no way arcane. They ask that one inquire diligently for evidence, proportion belief to it, draw inferences from it carefully, accept the conclusion and act accordingly. An essential component is preparedness to accept the defeasibility of belief, which means being ready to revise or abandon beliefs in the light of countervailing evidence if it comes. This - the acceptance of contingency - is one of the things people find hardest to do.

Most desire closure and neatness in their view of things, this being another impulse to religion, which provides the appearance of definite answers to everything from creation to the ultimate destiny of individuals.

The reason for there being so few "good believers" in the sense supplied by the ethics of rationality is that the vast majority of beliefs people hold are neither the outcome of rational inquiry, nor ever subjected to it. They were instead either acquired in circumstances where there was no question of their being evaluated critically, as in childhood, or they are the servants of their possessors' needs. And usually they are both. When this applies to the folk beliefs about how the world and other people work, the result is in the main a convenient one; but note that much science is the result of examining those beliefs, frequently finding better and richer explanations of why they work, and equally frequently going on to discover a great deal of interesting and useful matter besides.

There are four standard ways to explain religious belief. One is that it reveals the origin of the universe, why it works as it does (including, especially, the mysterious things that happen in it), and why it is full of evil and suffering. A second is that religion gives comfort and succour, provides a means (by prayer and sacrifice) to influence one's lot in this life and offers hope of a posthumous existence after this life has ended. A third is that religion ensures social order by promoting morality and social cohesion. And a fourth is that it rests on the innate gullibility of mankind, to say nothing of his ignorance and stupidity.

É«ºÐÖ±²¥

Religious people of course say that none of these explanations is correct, claiming instead that religion exists because there is a deity or deities.

Yet others offer theories drawn from sociobiology and from developmental and cognitive psychology. One such strand says that it is a human evolutionary adaptation to be hyper-credulous in childhood, this being the means by which children rapidly accept so many of humanity's fundamental notions (those useful and important folk beliefs again). But because religious beliefs are inculcated along with them, and receive constant reinforcement from the adults of the community, they acquire factual status.

It has also been hypothesised that consciousness of something like a "superego" has evolved to ensure behaviour that supports hierarchy, deference to those higher in the pecking order and loyalty to the tribe or clan. This structuring of relationships is recognised in all social mammals, and in humans extends to the idea of rulers in the sky.

Neurophysiology and cognitive psychology invoke understanding of the way the brain works to explain belief in supernatural phenomena. Complex neurological systems process information and draw inferences. Most of this work is conducted far from scrutiny by the conscious mind in specialist subsystems at deep levels of brain structure. The brain is hard wired to classify things into five "ontological categories" - animals, plants, tools, natural objects and persons - about which its subroutines form expectations and construct hypotheses, and from which they draw inferences.

Together, this allows the brain to represent the world to itself, and to interpret it.

Given this empirically well-grounded account of brain functioning, together with an associated account of the mind's psychological organisation, nothing needs to be invoked from outside these resources to explain how people can acquire belief in the supernatural. Because minds can remove concepts from their normal settings while keeping most of their standard inferential connections, they can easily picture ghosts and gods to themselves. For example, by combining the idea of something invisible (such as the air) with the idea of something that can secretly hear and see what we do (such as a person spying on us), we can easily devise the notion of a god, an angel or a ghost. A ghost is a creature that cannot be seen and can pass through solid objects such as walls - and thus differs from ordinary creatures in the natural world - but it can hear and see what we do, and interact with us in other ways, thus retaining significant natural characteristics of animals and persons. And then the mind stops: it does not ask how the combined aspects can be consistent with each other or what mechanisms respectively explain them.

These no doubt correct thoughts from cognitive science offer an explanation of how concepts of the supernatural easily arise in uncritical minds. But they leave to sociology (and psychopathology) the task of answering questions about why people need religious beliefs, why death is such a recurrent anxiety in them, why ritual matters so much to them, and why they give rise to fundamentalism and violence.

The four standard answers listed earlier are consistent with the cognitive science account, but go further by addressing these latter questions. They prompt an important reminder: that religion is a cultural matter as well as an artefact of the way our brains work, and culture is not merely an epiphenomenal outcome of neuronal activity, but in significant respects is the result of feedback from the social environment it has itself created.

To say this is to recognise that culture has an existence independent of individuals (though not of the collective they constitute). It also affects them as much as they affect it. Most people are passive spectators of culture, but significant minorities among them crucially influence its development and content, chief among them religious leaders, demagogues, writers and thinkers. They influence the character of the belief systems constituting the culture by endowing them with highly articulated literatures and traditions, liturgies and myths.

In the broadest sense, aspects of philosophy and the arts therefore count as a form of "cultural criticism" when they question and challenge the beliefs making up those systems. In that guise they are forms of "good believing", especially when they apply the criteria of the ethics of rationality to whatever happens to be currently accepted - sacred cows, shibboleths, cant, rhetoric, spin, newspeak - and to vestment-and-incense-disguised nonsense that goes echoing in the beauty of plainsong down ancient naves.

One of the ways this challenge can work is by refusing to be silenced before the defensive outworks that belief systems typically erect around them. In the case of religion, there are two that are especially noteworthy. One is the idea of "sacredness" - a given "holy" book is "sacred" and to mistreat the stack of paper from which an individual copy of it is made is "sacrilege" and "profanation". Likewise, to ridicule a belief, or to deny a claim to special treatment or exemption on the grounds of membership to a given religion, is to cause "offence". This latter is a cheap and easy move; subscribers to various religions are quick to lay claim to it - Christians are offended by the musical Jerry Springer - The Opera , Sikhs by Gurpreet Bhatti's play Beshti, Muslims by Salman Rushdie's The Satanic Verses . This would be acceptable if those offended by religious beliefs and practices could, just on that ground, close down churches and mosques in return.

But there is the rub: no rational individual could fail to see the point of the liberal principle that people should be allowed to believe what they like, in private and provided it does no harm to others, because there is no good argument against that freedom (even if there is every good argument against the belief thus privately entertained - whether in fairies or gods) and many excellent arguments for it.

Responsible intellectual endeavour consists in finding and maintaining the balance between two essential virtues: open-mindedness and critical scepticism. Scientists are the least dogmatic of inquirers, because the premise of their enterprise is that their best current theories might have to be revised or rejected if contrary evidence turns up. They try to make the strongest possible case for their theories, but they subject it to the relentless scrutiny of their peers. In that open exchange of claim and critical assessment lies the progressive nature of what they do.

What a far cry from religion. As the paradigm of responsible belief-formation, science is open, tentative and always subject to test.

Religion is dogmatic, final, closed, knows all the answers, damns as a heretic anyone who disagrees and too frequently kills them to boot.

There are aspects of serious science that are open to relatively few, because they require capacities that are not within everyone's reach or to everyone's taste, such as a competency in mathematics and the ability to see things in counterintuitive ways. But one does not have to be a scientist to be a "good believer"; that is open to all of us, if we would make the effort and adhere strictly to the requirement of intellectual honesty. Our world would quickly become a very different place if the majority of its inhabitants made a sincere effort to proportion belief to evidence, and to recognise that any belief they hold might be false.

A. C. Grayling is professor of philosophy at Birkbeck, University of London. His book, Descartes: The Life of Ren¨¦ Descartes and its Place in his Times , is published by Simon and Schuster, ?20.00.

É«ºÐÖ±²¥