RSS 2.0 Feed

» Welcome Guest Log In :: Register

Pages: (4) < [1] 2 3 4 >   
  Topic: BIO-Complexity, the shiny new ID journal< Next Oldest | Next Newest >  

Posts: 387
Joined: Aug. 2009

(Permalink) Posted: April 16 2012,07:38   

Quote (Wesley R. Elsberry @ April 07 2012,18:26)
It's the usual horse-puckey, I see. Dembski et al. cite me and Jeff Shallit to say that they looked at more genetic algorithms than just stuff like "weasel" that obviously has the target in it. They do this to say that Dave Thomas was wrong in saying:


They claim that GAs cannot generate true novelty
and that all such “answers” are surreptitiously introduced into the program via the algorithm’s fitness
testing function

They take issue with a quote Thomas makes, pointing out just how specific that quote was and how general Thomas' claim was.

But one can justify Thomas' claim for Dembski at least, given the following:


This result refutes the claim that evolutionary algorithms can generate specified complexity, for it means that they can yield specified complexity only if such algorithms along with their fitness functions are carefully adapted to the complex specified targets they are meant to attain. In other words, all the specified complexity we get out of an evolutionary algorithm has first to be put into the construction of the evolutionary algorithm and into the fitness function that guides the algorithm. Evolutionary algorithms therefore do not generate or create specified complexity, but merely harness already existing specified complexity. Like a bump under a rug, the specified complexity problem has been shifted around, but it has not been eliminated.

There are no caveats there about multi-part, complex systems or what-have-you; just a straight-up universal claim about the abilities of evolutionary computation. Given that Dembski hadn't at that point even gotten well onto the dodge of claiming that CSI only meant CSI above his "universal improbability bound", this can only be taken to mean that he intended it to apply even to measures of "local small probability" as discussed in "The Design Inference".

So how does that work if you are a theistic evolutionist - All living things we see today are the result of evolutionary processes acting on, and from, the first life forms - but none of the specified complexity we see in life today is a result of those processes.

  102 replies since May 13 2010,21:56 < Next Oldest | Next Newest >  

Pages: (4) < [1] 2 3 4 >   

Track this topic Email this topic Print this topic

[ Read the Board Rules ] | [Useful Links] | [Evolving Designs]