withallyourmind.net

In two previous blogs I have examined the first two of Moreland’s three formulations of the design argument. It uses an approach to evidence evaluation called Bayes’ theorem:

I first remember running across this theorem in medical school at Penn State. It provides a means for evaluating the probability of something as being true based on your prior probability (before the evidence was obtained) and in light of the new evidence. I use this principle in everyday decision making in medical practice. I have a prior probability in my mind of how likely I think it is that someone has a certain disease process. After obtaining the results of a specific test, I then have a post-test probability. To look more concretely at the theorem, here is how the terms are defined. It’s worth the few moments required to make sense of the above equation.

P (T/E) = the posterior probability of T given evidence E.
P (E/T) = the likelihood of E given T (i.e., If we accept T, does that make E certain, very likely, or just plausible?)
P (T) = the prior probability that T is true apart from evidence E.
P (E) = the probability that E will obtain apart from T, also called the expectedness of E.

So, we read the theorem this way: The probability of T being true given evidence E is equal to the prior probability of T being true without evidence E, multiplied by the likelihood that E will occur given that T is true, divided by the prior probability that E is expected to occur without T being true. If the probability that T is true given E is greater than the probability that T is true without E, P (T/E) > P (T), then the evidence E offers positive support for T.

Okay. What does all this have to do with the design argument? Let’s let T be the hypothesis that their is a theistic designer of the universe. Let E be the appearance of design found in the universe. There are three factors in the Bayesian theorem. P (E/T) is the probability that design would occur in the universe given that God was the designer, and this would be equivalent to 1, P(E/T) = 1, in Christian theism. This would be multiplied by the probability that a theistic designer exists apart from the evidence of design (e.g., based on other arguments for God’s existence). This would then be divided by the likelihood that the universe would have the appearance of design if God did not exist.

If P (E/T) is equal to 1, then it disappears from the equation. One could then argue that P(T) is not insignificant because of other arguments, and also argue that the likelihood of the appearance of design without the existence of God P (E) is quite low. A large number divided by a small number yields a positive posterior probability that a theistic designer exists.

In his introduction to The Creation Hypothesis, Moreland next examines criticisms of the design argument. I will look at those criticisms and Moreland’s response beginning in the next blog.