Opinions on the future of religion vary, especially on the future of religion in the West. In days of yore, which is to say the 19th century, the standard academic view was that science had bested religion as the explanation of events in the physical world. Why did it rain? Because of wholly (not "holy") meteorological factors, not because of any direct action by God. God did not dump buckets of water from heaven on to a chosen spot below. No longer was God even the indirect source of rain. If God taken as the creator of the laws of meteorology could not be ruled out, there was scant reason to rule God in.

Religion was not thereby eliminated from everyday life. On the contrary, religion could still serve any of the many functions not fulfilled by science. Where science could only explain, religion could prescribe - ethics. Jesus was now to be seen not as a miracle worker, whose deeds were intended to be contrary to scientifically explicable events, but as the ideal human, who practised what in the Sermon on the Mount he preached.

In the 20th century, religion was seen - again, by academics - as never having in fact been in competition with science. Now science was to be thanked for having forced believers to recognise belatedly that religion had never been an explanation of events in the physical world. Rather, religion had always been almost anything but an explanation of why rain falls. Science made a virtue out of a necessity: saving religion from eternal cultural purgatory.

What was, then, the proper function of religion? There was no single answer. Religion might be said to serve to unite a group, to pass on tradition, to spur an encounter with the unconscious, or to make life meaningful. Religion could serve many functions at once, as long as none of them involved explaining the world around us or, even worse, controlling the world around us by impelling the rain to fall.

ÖйúAƬ

Many of these non-explanatory functions were ones of which adherents were doubtless unaware. "So what?" was the rejoinder. Most individuals do not know what makes them tick physically. Why should they know any better what makes them tick religiously?

There thereby arose a divide between the insider's and the outsider's point of view. The insider is the adherent. The outsider is the scholar, which usually means the social scientist - sociologist, anthropologist, psychologist, economist or political scientist. The scholar may also be a practitioner, but the positions are distinct, and even when they are compat-ible, as they may or may not be.

ÖйúAƬ

The division between insider and outsider goes beyond that between recognised functions and unrecognised ones. The outsider often is a comparativist, which usually means a theorist. The function of religion for people X is assumed to be the same as that for the religion of people Y. The religion of people X is studied as simply a case in point, with the differences between it and the religion of people Y demoted to mere details. That for members of each religion the "details" are more than trivial represents just one more divide between the approaches.

Both 19th- and 20th-century theorists assumed that religion taken as explanation is incompatible with science. But that assumption is based on intellectual incompatibility, not on any survey of adherents. If adherents, who as modern Westerners are assumed to accept science, nevertheless continue to pray to God to get the crops to grow, to cure illness and to fend off death, they are scorned as simply inconsistent. They have failed to recognise the logical incompatibility of their religion with their science. It is not theorists who have failed to recognise what religion means for them.

For example, if a plane crash is attributed to bad weather but the survival of a handful credited to the grace of God, then survivors or their kin are considered too overwrought to be able to face up to the incompatibility of their religion with science. In other words, adherents are not really religious, for logically they cannot be.

Those who survived did so for scientific reasons, such as the location of their seats. That survivors or relatives themselves thank God for their survival is of no consequence. Again, there is a divide between insiders and outsiders. Like doctors vis-¨¤-vis patients, outsiders know better.

The parallel between believers and patients captures the hiatus between adherents and theorists. True, adherents have the religion. But then patients have the disease.

Adherents, like patients, are mere informants, or sources of information. Furthermore, the information that they have is usually inadequate for the equivalent of a diagnosis. Theorists are experts. They know more about religion than adherents. Specialists in individual religions know more about those religions than practitioners do. Believers no more have final authority on their religiosity than patients do on their illness.

ÖйúAƬ

How does this divide between believers and scholars bear on the issue of the future of religion? The answer is that the state of religion is not to be determined by polls or interviews. For decades, scholars themselves have debated the religiosity of contemporary Westerners. The reigning view had been that Westerners are becoming ever more secular and that religion is dying out. But this view has been questioned. At present, there is a stand-off: Western Europe is deemed ever more secular, but the US, not to mention Latin America, is deemed ever more religious. So each side can claim half-victory.

What counts is that the determination of religiosity is based on the equivalent in medicine of presenting symptoms. Do contemporary Westerners profess belief in God? Do they pray? Do they go to church or synagogue? Do they behave in accordance with their professed convictions? The more they do, the more religious they are.

ÖйúAƬ

But this approach, even though enlisted by outsiders, defers to the insider. Twentieth-century theories of religion do not, and in several ways.

First, religion can, again, be serving functions to which adherents are oblivious. If, as some sociologists assert, religion functions to unite a group, then even token attendance at worship may spell more religiosity than adherents recognise. Seemingly profane religious gatherings, such as picnics and sports events, may be far more religious than participants fathom. If, as some psychologists maintain, religion functions to provide an encounter with the believer's unconscious, then by definition the believer does not know that even perfunctory expressions of dependence on God or pleas for forgiveness may contain deep feelings towards a parent, of which God would be an unconscious symbol.

Second, there can be cases of religion unrecognised by the very persons belonging to them. If religion functions to unify a group, then football supporters may constitute a religion of their own. After all, they identify themselves as followers of a team, and they pit themselves against followers of any rival team.

They "idolise" and "worship" the players. They take for granted a gap between themselves and the players, the best of whom are called "gods". Fans seek to get close to the players but never presume to efface the line between themselves, who are mere mortals, and the players. A match is a religious gathering, with its own agenda and decorum. One goes to the stadium to participate in the worship of the team. One feels saved when the team wins, damned when the team loses.

Celebrities in many domains of life are treated as gods, and fans are their worshippers. Celebrities are seen as superhuman, not in all respects but, like gods in many religions, in a single respect. Zeus was strong, Aphrodite beautiful. The Hebrew God is neither omnipotent nor omniscient. Celebrities live like gods and get away with what ordinary folks do not. It could, then, be argued that fans create a religion of their own.

Yet unconventional religiosity is not confined to hero "worship". Any cause to which persons unite in their devotion can be called a religion. Nationalism, communism, the saving of the planet (Gaia worship!) and even professions such as science and medicine can garner as much devotion as any conventional religion. Those willing to give their lives, literally or figuratively, to their causes may stand more committed than practitioners of standard religions. Unconventional religions can have a pantheon, an officialdom, an orthodoxy, rituals and ethics as systematic as those of any standard religion. Gatherings can be as intense as those in an evangelical service. Participants can, in short, be characterised as fervently religious, and even when, in ordinary terms, they proudly consider themselves atheists. Once again, the outsider trumps the insider.

ÖйúAƬ

Has there been a present-day return to secularism? My answer should be obvious: it all depends on what you mean by religion.