Even William Dembski cannot explain CSI in any meaningful way (See Dembski 2005). First it’s a probability, then it’s an upper bound on probability, but he still uses it when the probability is great then 1.0 (which is nonsensical). Dembski calls it an Information Measure, but it is not. The CSI value is -2*log(p) where p is supposed to be a probability but isn’t. Ignoring those problems and a host of more serious troubles (See Elsberry and Shallit 2011, and Devine 2014), a high value of CSI indicates simplicity not complexity - the opposite of his claim. Dembski uses Kolmogorov Information as a measure of complexity, but the scale is backwards: Low KI means high CSI and vice versa. Either way, Dembski never establishes a connection between this measure and biological complexity, and never published corrections.
5 Likes