The Argument from Bayes' Theorem
A. The theorem enables to calculate the posterior probability of a hypothesis, given some new
piece of evidence: P(h/e).
P(h/e) = P(h)P(e/h)/P(e)
B. We need three inputs: (1) the prior probability of h, P(h), (2) the likelihood of the evidence,
assuming the truth of the hypothesis, P(e/h), (3) the prior probability of the evidence. For P(h/e)
to be high, we want P(h) and P(e/h) to be fairly high, and P(e) to be very low.
C. We can compute the probability of P(e) by the formula:
P(e) = P(e/h)P(h) + P(e/h)P(h)
Since we want P(e/h) to be high, P(e/h) must be very low. That is, the evidence e must be
something that would be very unlikely (surprising) if the hypothesis were false.
D. If P(e/h) > P(e/h), and P(h) ’ 0, then we say that the evidence e confirms hypothesis h. That
is, e raises the probability of h: P(h/e) > P(h), the posterior probability of h is higher than the
prior probability of h.
E. Swinburne's h = the existence of the God of classical theism (an infinite, uncaused, perfectly
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '09
 JorgeRigol
 Intelligent design, Prior probability, A. Dembski

Click to edit the document details