From: Glenn Morton (glenn.morton@btinternet.com)
Date: Fri Nov 22 2002 - 13:37:49 EST
Josh Bembenek writes:
>-At best you've delineated a situation where the filter will give a false
>negative, or a non-design reference to something designed. I
>think atheists
>and critics should be and are much more worried about false positives.
>Along these lines anything the filter calls "designed" should be seen as
>non-design structures for some reason. Who cares if the filter can't pick
>up some cryptic sequence or not?
Because this back door shows why Dembski's method fails miserably. The point
is that if any random looking sequence MIGHT be a designed sequence, Dembski
can't claim ANY sequence is not designed. In other words, he has no proof
that a sequence is merely random and due to chance. Thus, after going
through Dembski's method, all sequences are still candidates for design. I
repeat ALL sequences are candidates for design. Thus, Dembski hasn't
discriminated any of them with any certainty at all. And I would say that
those which he declares as due to design (like this very sentence) he knows
to be designed because he reads English, NOT because of his supposed
methodology. Thus, in the end, his methodology adds absolutely NOTHING to
the discussion.
glenn
see http://www.glenn.morton.btinternet.co.uk/dmd.htm
for lots of creation/evolution information
anthropology/geology/paleontology/theology\
personal stories of struggle
You are arguing that the filter cannot
>discriminate between designed and non-designed sequences among the
>class of
>sequences that the filter will give a non-design inference for.
>Fine, some
>designed structures will not be recognized. However, what we really care
>about is what to make of the sequences that are given a positive design
>inference. The question is whether we can positively assert that
>something
>like the flagellum bears the features of design.
How do you know that it was designed. The very issue is whether it was
designed or evolved (not ruling out design via evolutionary processes which
are now in vogue in engineering). You are assuming the consequence here. It
looks designed to you but you have no objective, mathematical methodology
for stating that. If you do, show the math in your reply. If you can't, then
you have not proven anything.
For the
>flagellum, we know
>the code, the translation and the final product, we don't need to discuss
>cryptic messages. If the flagellum is not designed, the only options for
>the filter is for it to actually give indications of non-design or give a
>false positive.
OK, use this methodology and lay out the details here for this audience how
the method reaches the decision. Show the math, not the words.
Conversely a flagellum is designed, the filter can only
>readout design or give a false negative. False negatives are
>irrelevant to
>critics, because they don't want to see biological design and falsely
>calling a design structure not designed suits them well. Dembski,
>however,
>is worried about false negatives because he is trying to detect design in
>nature. Only for those objects for which design is detected by the filter
>should we be interested in further exploring whether or not they
>are really
>designed. As for true or false negatives, those issues can be explored
>later, but the fact that you have identified what you believe to
>be a false
>negative doesn't invalidate the approach.
It does invalidate the approach IF every sequence he analyzes still has the
possibility of being designed. There is no ability to distinguish by use of
the methodology.
This archive was generated by hypermail 2.1.4 : Sat Nov 23 2002 - 00:37:58 EST