With religious claims, compatibility won't do

The article is a little old now – too old by most standards – but Julian Baggini's post in the 'Comment is Free' section of the Guardian is excellent:

The religious believer could bite the bullet, accept that religion does make some empirical claims, and then defend their compatibility with science one by one. But the fact that two beliefs are compatible with each other is the most minimal test of their reasonableness imaginable. All sorts of outlandish beliefs – that the Apollo moon landings never happened, for instance – are compatible with science, but that hardly makes them credible. What really counts, what should really make the difference between assent and rejection of an empirical claim, is not whether it is compatible with science, but whether an evidence-led, rational examination of a view supports it better than competing alternatives.

Keith Ward submits otherwise; Jerry Coyne objects to Ward's argument:

Saying that God created the universe is no more a “perfectly acceptable factual claim” than is “the universe was created by a giant turtle” or “invisible and undetectable fairies move the pistons of my car.” A factual claim is “acceptable” when it is both testable and doesn’t violently contradict what we know of the world. And if “no known scientific technique” can answer the question of whether some deistic act ultimately stated the universe, then, contra Ward, there is no “rational” way to answer such a question.

Ophelia Benson and Jim Houston also weigh in.