Science and Religion Redux
The issue is of course largely semantic, on what should we understand by "religion". Provisionally we can define as a religious doctrine any one that includes reference to a supernatural realm. (While this may leave out some forms of religion, for example pantheistic ones, it probably reflects quite well the ordinary Western use of the term). It is clear then that a religious claim will always have a metaphysical element, but also that it need not be purely metaphysical. Some religious statements are so (e.g. "God exists"), while some have in addition an ethical content ("Pride is a sin against God") or an empirical content ("God created the world in six days").
Stephen Jay Gould introduced the concept of NOMA (Nonoverlapping Magisteria) for the idea that proper religious claims should not address empirical matters. (The original article can be read here). This seems at first glance to make a nice division, that many people in our cultural climate would endorse: the empirical and testable goes to science, the metaphysical and ethical to religion. But things can't be so simple.
To start with, a way of using NOMA which one sees often, especially by religious people (for example in the context of bioethics), is along the lines of "Science can tell us how to do something, but only religion can tell us if we should do it". The fallacy of the binary division is clear here: it is true that science cannot answer ethical questions, but there is no reason to think that only religion can do so (as opposed to nonreligious ethical philosophy, or even unsystematized moral feelings). The situation is not symmetrical: there is every reason to trust the scientist more than the priest on the question of how a given contraception technique works, but there is no solid reason to trust the priest more than the scientist, the philosopher or the man of the street on the question of whether to use it is moral. This, however, could be said to be a mistake in popular use of NOMA rather than in the principle itself, which should state only that "religion does not deal with empirical matters, only with metaphysical or ethical ones" without saying that only religion can make ethical pronouncements (or metaphysical; there is nonreligious metaphysics, after all).
Another criticism of NOMA, that Jason makes in his post, it that it can be seen as commiting the No True Scotsman fallacy. One can imagine the defender of NOMA, when confronted with a creationist who makes lots of empirical claims based on his religion (about the young age of the Earth, the Flood, etc.), answering: "Your position is not really religious, because true religion is not concerned with empirical matters". This would seem to be a cheap redefinition that goes not only against the popular use of the term, as stated at the beginning of this post, but also against the creationist own subjective religious conviction.
However, Brandon in his post defends NOMA against this charge (though critizizing other aspects of it) saying that what NOMA aims at is clarifying a conceptual confusion. Perhaps in popular use the term "religion" applies to some empirical doctrines; but if we are convinced that there is a deep divergence between empirical and nonempirical claims, and that only science has authority to make the first ones, and that the core and fundamental elements of religion are not empirical, then the NOMA move would not be a cheap redefinition but one that aims at clearing a conceptual confusion and making for a more organized intellectual panorama.
But how true is it that the fundamental elements of religion are not empirical? I would like to think it is true, as it would eliminate all possible conflict between science and religion, and I think it is true of a pure, idealized form of religion. But there is hardly any real-world religious tradition for which this is true. Take Chistianity for example; surely a core element of it is the claim that a man called Jesus lived 2000 years ago, was crucified, and resurrected. What if we found convincing historical evidence that Jesus never existed, or that he existed but did not die in the cross? While I don't think it is likely that this will happen (unlike, for the latter example, the many amazingly guillible people who are taking seriously The Da Vinci Code) it is surely an empirical question to be decided in principle by science. (Even the resurrection is: concievably we could find Jesus' tomb with his remains still there, disproving a literal bodily resurrection.) What about the idea of an inmaterial soul? Suppose science comes to give a complete and convincing explanation of human consciousness that does not involve the spirit; wouldn’t that force a serious reconsideration of some core religious doctrines?
So, while I sympathize with NOMA as stating an ideal for religion, I think it stands in too large contradiction with our central concrete examples of religion to be "true", even in the way Brandon suggests. Insisting on it could be a useful move for making religion evolve in a good direction, but not a clarification of a minor conceptual confusion that does not affect fundamental religious tenets.
My ideal of religion would be one that had only purely trascendental and ethical doctrines, basing them on faith but claiming no rational authority, and tolerating other religious and nonreligious (for example philosophical) views on its subject. Its function would be to provide a way for people to give meaning to their lifes in a trascendental context for those who need to. I personally would not adhere to it; I am too much of a rationalist to be able to just have faith in a non-rationally based idea, and I don’t feel the need of a trascendental meaning for my life. But I would have no quarrel at all with those who embraced it, and could even view them with sympathy.
Sadly, this kind of religion is quite far away from those that have the spotlight in the real world.