As a young teenager in proudly peaceable Canada during the romantic 1960s, I was a true believer in Bakunin's anarchism. I laughed off my parents' argument that if the government ever laid down its arms all hell would break loose. Our competing predictions were put to the test at 8:00 A.M. on October 17, 1969, when the Montreal police went on strike. By 11:20 A.M. the first bank was robbed. By noon most downtown stores had closed because of looting. Within a few more hours, taxi drivers burned down the garage of a limousine service that competed with them for airport customers, a rooftop sniper killed a provincial police officer, rioters broke into several hotels and restaurants, and a doctor slew a burglar in his suburban home. By the end of the day, six banks had been robbed, a hundred shops had been looted, twelve fires had been set, forty carloads of storefront glass had been broken, and three million dollars in property damage had been inflicted, before city authorities had to call in the army and, of course, the Mounties to restore order. This decisive empirical test left my politics in tatters.
Perhaps I, too, am a Pollyanna to believe that people would remain good when unobserved and unpoliced by God. On the other hand, the majority of the population of Montreal presumably believed in God. Why didn't the fear of God restrain them when earthly policemen were temporarily removed from the scene? Wasn't the Montreal strike a pretty good natural experiment to test the hypothesis that belief in God makes us good? Or did the cynic H. L. Mencken get it right when he tartly observed: 'People say we need religion when what they really mean is we need police.'
Obviously, not everybody in Montreal behaved badly as soon as the police were off the scene. It would be interesting to know whether there was any statistical tendency, however slight, for religious believers to loot and destroy less than unbelievers. My uninformed prediction would have been opposite. It is often cynically said that there are no atheists in foxholes. I'm inclined to suspect (with some evidence, although it may be simplistic to draw conclusions from it) that there are very few atheists in prisons. I am not necessarily claiming that atheism increases morality, although humanism — the ethical system that often goes with atheism — probably does. Another good possibility is that atheism is correlated with some third factor, such as higher education, intelligence or reflectiveness, which might counteract criminal impulses. Such research evidence as there is certainly doesn't support the common view that religiosity is positively correlated with morality. Correlational evidence is never conclusive, but the following data, described by Sam Harris in his Letter to a Christian Nation, are nevertheless striking.
While political party affiliation in the United States is not a perfect indicator of religiosity, it is no secret that the 'red [Republican] states' are primarily red due to the overwhelming political influence of conservative Christians. If there were a strong correlation between Christian conservatism and societal health, we might expect to see some sign of it in red-state America. We don't. Of the twenty-five cities with the lowest rates of violent crime, 62 percent are in 'blue' [Democrat] states, and 38 percent are in 'red' [Republican] states. Of the twenty-five most dangerous cities, 76 percent are in red states, and 24 percent are in blue states. In fact, three of the five most dangerous cities in the U.S. are in the pious state of Texas. The twelve states with the highest rates of burglary are red. Twenty-four of the twenty-nine states with the highest rates of theft are red. Of the twenty-two states with the highest rates of murder, seventeen are red.*)
Systematic research if anything tends to support such correlational data. Dan Dennett, in Breaking the Spell, sardonically comments, not on Harris's book in particular, but on such studies generally:
Needless to say, these results strike so hard at the standard claims of greater moral virtue among the religious that there has been a considerable surge of further research initiated by religious organizations attempting to refute them. one thing we can be sure of is that if there is a significant positive relationship between moral behaviour and religious affiliation, practice, or belief, it will soon be discovered, since so many religious organizations are eager to confirm their traditional beliefs about this scientifically. (They are quite impressed with the truth-finding power of science when it supports what they already believe.) Every month that passes without such a demonstration underlines the suspicion that it just isn't so.
Most thoughtful people would agree that morality in the absence of policing is somehow more truly moral than the kind of false morality that vanishes as soon as the police go on strike or the spy camera is switched off, whether the spy camera is a real one monitored in the police station or an imaginary one in heaven. But it is perhaps unfair to interpret the question 'If there is no God, why bother to be good?' in such a cynical way.*) A religious thinker could offer a more genuinely moral interpretation, along the lines of the following statement from an imaginary apologist. 'If you don't believe in God, you don't believe there are any absolute standards of morality. With the best will in the world you may intend to be a good person, but how do you decide what is good and what is bad? Only religion can ultimately provide your standards of good and evil. Without religion you have to make it up as you go along. That would be morality without a rule book: morality flying by the seat of its pants. If morality is merely a matter of choice, Hitler could claim to be moral by his own eugenically inspired standards, and all the atheist can do is make a personal choice to live by different lights. The Christian, the Jew or the Muslim, by contrast, can claim that evil has an absolute meaning, true for all time and in all places, according to which Hitler was absolutely evil'.
Even if it were true that we need God to be moral, it would of course not make God's existence more likely, merely more desirable (many people cannot tell the difference). But that is not the issue here. My imaginary religious apologist has no need to admit that sucking up to God is the religious motive for doing good. Rather, his claim is that, wherever the motive to be good comes from, without God there would be no standard for deciding what is good. We could each make up our own definition of the good, and behave accordingly. Moral principles that are based only upon religion (as opposed to, say, the 'golden rule', which is often associated with religions but can be derived from elsewhere) may be called absolutist. Good is good and bad is bad, and we don't mess around deciding particular cases by whether, for example, somebody suffers. My religious apologist would claim that only religion can provide a basis for deciding what is good.
Some philosophers, notably Kant, have tried to derive absolute morals from non-religious sources. Though a religious man himself, as was almost inevitable in his time*), Kant tried to base a morality on duty for duty's sake, rather than for God's. His famous categorical imperative enjoins us to 'act only on that maxim whereby thou canst at the same time will that it should become a universal law'. This works tidily for the example of telling lies. Imagine a world in which people told lies as a matter of principle, where lying was regarded as a good and moral thing to do. In such a world, lying itself would cease to have any meaning. Lying needs a presumption of truth for its very definition. If a moral principle is something we should wish everybody to follow, lying cannot be a moral principle because the principle itself would break down in meaninglessness. Lying, as a rule for life, is inherently unstable. More generally, selfishness, or free-riding parasitism on the goodwill of others, may work for me as a lone selfish individual and give me personal satisfaction. But I cannot wish that everybody would adopt selfish parasitism as a moral principle, if only because then I would have nobody to parasitize.
38.
Note that these colour conventions in America are exactly the opposite of those in Britain, where blue is the colour of the Conservative Party, and red, as in the rest of the world, is the colour traditionally associated with the political left.
39.
H. L. Mencken, again with characteristic cynicism, defined conscience as the inner voice that warns us that someone may be looking.
40.
This is the standard interpretation of Kant's views. However, the noted philosopher A. C. Grayling has plausibly argued (New Humanist, July-Aug. 2006) that, although Kant publicly went along with the religious conventions of his time, he was really an atheist.