The Secular Outpost
Follow Patheos Atheist:
Jeffery Jay Lowder is President Emeritus of Internet Infidels, Inc., which he co-founded in 1995. He is also co-editor of the book, The Empty Tomb: Jesus Beyond the Grave.
Jefferys’ review is an excellent exposé of an all-too-common genre of religious apologetic. Many theistic arguments take the form of claiming that, for some evidence E, some design hypothesis D, and some naturalistic hypothesis N, p(E/D) > p(E/N). But, of course, depending on what we take for E and D, p(E/D) can have the value of one, zero, any value between one and zero, or no definite value at all. The same holds for p(E/N). If we take E to be the fact that the fundamental physical constants fall within the complex-life permitting range, and D to be the minimal claim that the universe had some sort of an intelligent designer (e.g. God, the Q Continuum, a celestial supercomputer, or whatever), then, as Jefferys observes, p(E/D) will have no definite value. It could be one, zero, or anything between one and zero, depending on any number of imponderables. You simply have to specify a lot more about D before p(E/D) has any definite value.
On the other hand, as ID theorists not-so-secretly think, maybe D is a definite being, maybe in fact (guess who!) the theistic God. If D = T, where T is the hypothesis that the theistic God is the designer, then we could reasonably assume that p(E/T) is quite high. But then we have to ask about p(T), the background probability that of indefinitely many possible ultimate supernatural entities that might have existed (e.g. Platonic ideas, the Plotinian One, Casper the Friendly Demiurge, etc.), we would be so impossibly lucky as to get the theistic God. Seemingly, then, p(T) would be quite low. Worse, p(E/~T) might be rather high. ~T would be the conjunction of all those hypotheses that might account for E other than the theistic hypothesis. Thus, ~T would encompass any ultimate naturalistic hypotheses, as well as any hypotheses of a supernatural designer other than the theistic God. Hence, the total probability of E, p(T) X p(E/T) + p(~T) X p(E/~T), might be quite high. In that case, even if the likelihood p(E/T) approximates one, the probability of T given E, p(T/E) might be quite low, so the argument would fail.
Follow Patheos on