RSS 2.0 Feed

» Welcome Guest Log In :: Register

    
  Topic: Dembski, natural selection, & math, Analysis of IC and NS< Next Oldest | Next Newest >  
Wesley R. Elsberry



Posts: 4807
Joined: May 2002

(Permalink) Posted: Sep. 18 2002,03:38   

William Dembski has offered a very interesting piece concerning what he believes needs to be demonstrated to show that natural selection is a sufficient causal explanation for some system.  See it on the ISCID Brainstorms board.

The basic gist is relatively straightforward, though the notation looks a bit overblown.

Dembski starts with an analogy to demonstrating common ancestry of two lineages X and Y, whose initial states X_0 and Y_0 are actually the single common ancestor of all derived X_n and Y_n.   Dembski asserts:

Quote
In the best circumstance, each such X(i) and Y(j) must be explicitly exhibited and any arrows of causation connecting two organisms must produce small incremental changes that are highly probable on the basis of the Darwinian selection mechanism. The more intermediates that are missing from this picture and the more handwaving and just-so story-telling to describe the arrows of causation, the more problematic the evolutionary explanation.


It should be noted that common ancestry is not dependent upon the mechanism of natural selection being operative at every step, or indeed at any step.  Dembski's scenario completely ignores the evidence of molecular biology in applying sequence comparisons, which is largely based upon the evidence of the X_n and Y_n extant organisms, rather than X_0, Y_0, or any intermediates, since the molecular data from long extinct organisms is generally not available for anlaysis.  Nor does natural selection eschew incorporation of "large" changes, should such change have an adaptive advantage for the bearer.  It is well-developed in the evolutionary literature that "small" changes are more likely to have an adaptive advantage than "large" changes, and thus we should expect more "small" changes to be observed in lineages undergoing selection.  But that doesn't limit natural selection to *only* using "small" changes, as Dembski seems to imply above.  These departures Dembski takes from the biological reality of inferring common ancestry of lineages suggest that Dembski's approach is problematic.

As G.G. Simpson pointed out, intermediates are always missing, except where they are found.  Let me point out  what should count as a non-problematic example of common ancestry of two lineages, Globigerinoides trilobus and Orbulina universa.  A very good plate appears in the paper by Pearson et alia.

Quote
Pearson, P.N.; Shackleton, N.J.; and Hall, M.A., 1997.  Stable isotopic evidence for the sympatric divergence of _Globigerinoides_trilobus_ and _Orbulina_universa_ (planktonic foraminifera).  Journal of the Geological Society, London, v.154, p.295-302.


Figures from this paper are reproduced on the web in this page by Don Lindsay.

Pearson et alia adduce other evidence than what Dembski has offered in his argument.  They examine stable isotopic evidence to show that the divergence of G. trilobus into O. universa occurred sympatrically.  This also has a bearing on the sufficiency of the evolutionary account of common ancestry of these two species.  Notably, though, Pearson et alia do not invoke natural selection as the sole mechanism of change in this divergence.  The fact of the divergence is an issue separate from the underlying mechanism.

Dembski's underlying analogy for the remainder of his argument concerning the sufficiency of evolutionary explanations for IC systems excludes relevant classes of evidence, unnecessarily invokes a particular process as needed to be demonstrated, and ignores actual biological practice in showing common ancestry of lineages.  This, to say the least, is an inauspicious beginning for the remainder of his argument.

--------------
"You can't teach an old dogma new tricks." - Dorothy Parker

    
  0 replies since Sep. 18 2002,03:38 < Next Oldest | Next Newest >  

    


Track this topic Email this topic Print this topic

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