The political versions of religious nonbelief usually include affirmations of rationality in public policy. The recent Copenhagen Declaration on Religion in Public Life has “We submit that public policy should be informed by evidence and reason, not by dogma.”
All this assumes that there is a single agreed upon form of rationality. Maybe it also assumes that people who share in this form of rationality will, if well enough informed, converge upon a single policy. And it certainly seems to assume that we would be collectively better off if everyone were to behave rationally.
All of these assumptions seem dubious to me.
Rationality is notoriously difficult to pin down. Is it the rational utility-maximization of agents in economic theories? A deeper rationality that is capable of determining aims as well as letting us choose means? There are many proposed forms of rationality, and their demands can conflict with one another. If we work with less ambitious (and thus more reasonable) concepts of rationality, which allow a plurality of sometimes incommensurable sets of aims or ways of life, then we also cannot demand that all rational people must converge on some kind of agreement. In such situations, why should policy not be informed by any common forms of faith that may happen to exist, including religious traditions?
Moreover, rationality may not, collectively speaking, be all that good an idea. Rationality in individual choices can cause a society to collectively jump off a cliff. Some of this “irrational rationality” was easily observable in the recent global financial crisis. The incentives in many financial markets were such that financial actors had to follow the herd, even when it was clear that in a slightly longer term, disaster was extremely likely. So long term public policy, it would seem, should include mechanisms that work against individual rationality. Religion, with its often communitarian focus, and anti-rational thrust, might be just the sort of public policy instrument we need.
In any case, nonbelievers need some sort of argument that our form of rationality is a good idea, instead of treating it as some kind of self-evident fact.