> >Peter: I don't think the leading ID [intelligent design] people would be satisfied with an IC [irreducible complexity] definition admitting partial systems "not favored by selection", but viable. Rather, they would insist that each partial system on any conceivable evolutionary path would be lethal.
> I do not see how this could be proven. More importantly, I do not see prominent ID advocates promoting this approach. I do not see any connection between Dembski's specified complexity and the viability of intermediate systems. (For that matter, I do not think specified complexity is a very good correlate of special design). Many of Behe's complex systems are not necessary to life, so partial systems will not be inviable. Even some systems that currently must be complete for all modern organisms show traces of ancient gene duplications that indicate that an ancestral organism was getting by with fewer parts.
> Dr. David Campbell
In fact, it's Dembski, Behe or other ID people who should answer this. I
just answered you because I got the impression that you misrepresented
their position. I still think a viable partial system would necessarily
indicate that the whole system (of which it is a part) cannot be
irreducibly complex - by Behe's definition. Dembski's explanatory filter
involves 3 tests (W.A. Dembski, in: "Mere Creation" (InterVarsity Press,
1998), pp. 93 ff): (1) is the system highly probable (explanation: law)?
If not, (2) is it of intermediate probability (explanation: chance)? If
not, (3) is it specified in advance (explanation: design)? If not, it
happened by chance. If yes, it has specified complexity (according to
Dembski), or is irreducibly complex (according to Behe). If there is a
viable evolutionary precursor, the system doesn't pass test 2 of the
Dembski filter (but this precursor, in turn, may then have to be tested
for irreducible complexity).
You are right in doubting that irreducibility can be proven. You can
hardly ever prove that there cannot be any viable evolutionary precursor
- you just might find one some day. Negatives in this sense usually
cannot be proven. But all Dembski claims is inference to the most
Peter Ruest <email@example.com>
This archive was generated by hypermail 2b29 : Wed Nov 08 2000 - 11:08:16 EST