It has been commonplace in the history of moral theory to argue that having an obligation and being motivated to fulfill that obligation come apart. I have argued that this was the conception that Hobbes and Locke had. Each of the philosophers thought that we could have obligations (even in the state of nature) but that we needed, in addition to the obligation itself, some other motivating reason to fulfil the obligation. This can be seen as partly what a Kantian moral theory denies, in that they claim that the having of the obligation (or the recognition that one has it) is the only (legitimate) motivation to fulfil the obligation. So, if one has an anti-Kantian view of this sort one will have to appeal to some strong authority as an enforcer of the moral rules. Hobbes himself says that if there were a God then he would be the one to punish and reward those who break or follow the rules, but in his absence we need a strong Earthly authority.
It seems to me, though I admit that this is ultimately an empirical question, that belief in the existence of God and his willingness to punish and reward people who ignore or follow the dictates of morality is a strong motivator to obey said rules. It also seems to me that if people did not have a belief in God they would be more disposed to breaking the rules of morality when they were confident that they would not be caught by Earthly authorities (I mean, God is always watching, but the city of New York has its lapses). This is of course the problem of Hobbes’ intelligent Knave. Even if one is a Kantian about motivation (like I am), doesn’t one have to admit that fear of consequences has more motivational pull that does the recognition of obligation? Certainly not in all cases, but I mean generally among mankind.
Now, the utilitarian believes that the action (rule, preference, whatever) that promotes the greatest amount of happiness is the right action (rule, whatever) but our motivation for performance doesn’t matter. So, on utilitarian views one can do the right thing for the wrong reasons and still count as performing a moral action (though I sometimes think a Kantian has to say this as well). So, a world populated solely by atheists would be one that was less morally good than a world populated (mostly) by people who feared an all-powerful God. This is because, no matter how good the Earthly government’s enforcement of the moral rules is, it will not be 100% and so will not provide as much motivation to avoid immoral acts as belief that there is an all-powerful being who is always watching and judging you would. Given this it turns out that the utilitarian is obligated not only to avoid promoting belief in atheism, but also to promoting theism of a very strict sort.
Well, that wasn’t as short as I thought :)