In his leader column  for the New Statesman, Richard Dawkins makes a sharp distinction between religion as a cultural legacy -- as expressed in such cosy familiarities as Christmas carols and the King James Bible -- and religion as a living, and he would say pernicious, social force. The one, he thinks, is both traditional and "freely embraced by individuals". The other exercises "a real domination" over culture and politics.
Government, he maintains, "forces religion on society, in ways whose very familiarity disarms us" -- by which he means such things as bishops in Lords, the Charity Commission's assumption that religion is in the public interest, above all the inexorable spread of faith schools.
I can see where he's coming from. Indeed, I've written here before  about the anomaly of a growing "faith school" sector in an increasingly non-believing society. And, like him, I can relate strongly to the cultural aspects of traditional Christianity without believing a word of it. Nevertheless, I think he's missing something significant.
The cultural prestige of religion -- in this country, this means especially Anglican Christianity -- is what gives it the access to power, what buttresses its claim to speak on behalf of morality, what causes politicians to want to embrace it in the first place. The bishops' seats in the House of Lords, which would appear to be safe from proposed reforms to the composition of the upper house, are a legacy of history just as Handel's Messiah or school nativity plays are legacies of history.
The Christian bits of Christmas are no less Christian because they are cosy and familiar. As soon as you wish someone a "Merry Christmas" you are inviting anyone from a doorstepping evangelist to a Thought for the Day contributor to remind you that the "true meaning of Christmas" centres around the birth, allegedly in Bethlehem, from an inviolate virgin of the Saviour of Mankind.
Dawkins wants to have his Christmas pudding and eat it, I'm afraid. He complains about the United States, where the constitution separates church and state, that "rival religions" have long fostered a "tiresome" avoidance of the C word (except, that is, in Bing Crosby's "secular carol" White Christmas, which the good professor rightly abominates).
But the US is also a country in which no presidential candidate could ever admit to being an atheist or even agnostic, where the equation of religious belief with personal morality is accepted almost without question, where (according to a recent study ) atheists are considered less trustworthy than rapists.
The US might not have state-funded faith schools or bishops sitting ex officio in Congress, but religion is if anything even more deeply interwoven with culture over there than it is in Britain with its established church.
American politicians no less than British ones "believe in belief" -- nor is there the embarrassment factor that led Alastair Campbell to instruct Tony Blair not to "do God". If there's a difference (and of course there is) it's that while American politicians appear to believe in belief itself -- their own and other people's -- British politicians tend to confine themselves to stressing the social utility of religion.
Dawkins laments the "depressingly large number of intelligent and educated people", themselves perhaps non-believers, who "still vaguely presume without thinking about it that religious faith is somehow good for other people, good for society, good for public order, good for instilling morals, good for the common people."
There's nothing novel about such an assumption, of course. Politicians have been thinking along these lines since the days of the Roman Empire. As Edward Gibbon described the situation then, "The various modes of worship which prevailed in the Roman world, were all considered by the people, as equally true; by the philosopher, as equally false; and by the magistrate, as equally useful."
Is the assumption in any way justified?
Dawkins is right to point out that religion makes a poor philosophical foundation for morality. Fear of eternal damnation, as he says, is a "contemptibly immoral motive for being moral". And in fact research suggests  little difference between the basic moral intuitions of religious and non-religious people. (This is good news for believers, by the way, as it suggests that their moral behaviour is not, after all, the result of their fear of divine displeasure, but comes rather from the innate ethical sense they share with most normal human beings.)
Nevertheless, and even allowing for the peculiarities of particular creeds, religious teaching has tended to provide a good approximation of proper ethical reasoning -- good enough for the relationship between religion and morality to have become well-established over many centuries. Religion is not morality, but if its teachings had strayed too far from innate human moral sensibilities it's unlikely that it would have survived.
What we're seeing today, in fact, in issues such as gay rights or the role of women, is religion engaged in a game of moral catch-up with secular ethics. Only those that successfully adapt will have a secure long-term future. The process might almost be called Darwinian.
Equally Darwinian, of course, is the way in which religions have managed to inveigle themselves into people's ethnic and social identities in so obstinate a way that even as anti-religious a campaigner as Richard Dawkins, a man who believes fervently in disbelief, feels irrationally impelled to stand up and sing "O come all ye faithful".
Even he, it seems, has been infected by some version of the God meme.