PZ asks in comments whether "a disciplined adherence to a set of arbitrary rules" can be considered rational. But the whole point about Sayers’ argument, which I have never elsewhere seen so clearly put, is that she does not see the dogmas of Christianity as arbitrary. They correspond, she says, to the truth about human nature and the universe, and this correspondence becomes apparent when we test them. This isn’t, she says, an argument for the existence of God. What we say about God might simply be a way of describing the facts of human nature. But, either way, it would not be arbitrary. It could be tested – and it is.

This point matters because it strikes at what seems to me one of the central atheist misunderstandings of the religious, which is that their beliefs are arbitrary.

Sayers distinguished between the historical and theological assertions of Christianity. Theological assertions she regards as more testable, which, is counterintuitively true if they are assertions about the state of the universe, as she believed. The truth of an asserted one-off happening like the resurrection simply can’t be tested directly. In the end, we make up our minds by deciding how well it fits with the rest of the evidence. On the other hand, we can test the truth of an assertion like "to him that hath shall be given, but from him that hath not shall be taken away even that which he hath" or "it must needs be that offences come, but woe unto that man by whom the offence cometh" and experience suggests that these are very often true. Indeed, the second is an excellent short definition of tragedy.

I repeat, the point here is not that saying them proved that Jesus was God or anything like that; what these statements disprove is the Pharyngular assertion that religious talk is by its nature arbitrary and untestable.

I am aware of the countermove that says these things aren’t really religious if they are reasonable and testable. I just think it’s disgraceful and indeed anti-scientific. We don’t distinguish between essences and accidents in biology. Why do so in sociology, which is what the scientific study of religion comes down to? So the objection that theological statements are not really theological if they turn out to be statements about the universe seems to me arbitrary and unwarranted. You have to ask first how these statements are intended, and, at least in the tradition of European Catholic philosophy, they seem to be intended and understood as statements of fact about the universe.

This is, I think, what Cardinal Schönborn was saying when he denied that the argument for a designer was a matter of faith and said it derived from a philosophical truth. But Sayers puts the point better. Her own words will be in the next post.