Behe made the claims about unselected mutations, and in his responses to 
his critics in 2000 he claimed that well matched, and the order and 
arrangement of mutations would make the system his type of IC.
> 
> More to the point, my impression is Rusbult has conveniently forgotten
> the reason why Behe invented IC in the first place, as an alleged
> disproof of naturalistic evolution.  IMO the important thing is not to
> show that IC by any definition does or does not exist, but instead to
> show that, if that kind of IC does exist, it could not have evolved by
> unguided natural processes, or alternately, that it must have involved
> a purposeful intelligent agent.
Behe doesn't like to admit it, but his changes have been trying to 
maintain the IC stupidity, while keeping IC something that could not 
have evolved by natural mechanisms.  The stupid "scientific test" for IC 
that both Minnich and Behe put up in their Dover testimony was to start 
with a bacteria with out a flagellum and get it to evolve a flagellum. 
If a flagellum could have been evolved then Behe would be wrong and IC 
will have failed.  Both Minnich and Behe admitted that they had never 
attempted the only scientific test that they proposed, so it was obvious 
that neither cared about IC being testable.
> 
> But Rusbult explicitly acknowledges these new definitions of IC
> recognize the evolutionary possibilities of Behe's so-called indirect
> methods, where the parts change function over time, and where
> unnecessary parts are lost over time.  By so doing, these new
> definitions widen the definition of IC, but also lose their ability to
> identify what could not evolve, or what must have involved an
> intelligent agent, and by so doing, render IC moot.
> 
Behe as always admitted that biological evolution is a fact of nature. 
He has admitted that his designer might be dead because he doesn't see 
any evidence for designer activity in the last couple of hundred million 
years of biological evolution.  His last IC system evolved in 
vertebrates over 400 million years ago (the adaptive immune system). 
His flagellum evolved in bacteria both Archea and Eubacteria over a 
billion years ago.  That is the major reason why Behe can't tell anyone 
if the flagellum is his type of IC.  It evolved so long ago that much of 
the molecular history has been destroyed by multiple mutations a pretty 
much ever site in the proteins that can change.
Behe relies on biological evolution to help him demonstrate the order 
and arrangement of the mutations that will make a system his type of IC. 
  He thinks that there are orders and arrangements of mutations that 
would need to be done by his designer.  His crowing about finding two 
unselected steps in the evolution of systems demonstrates how he thinks 
that he will be vindicated.  He admits that two are not enough, but he 
claims that this is the edge of IDiocy, and that biological evolution 
should not be able to go beyond that.  Just like any other science 
researchers have just determined what happened and it isn't what Behe 
needs.  Labs like Thornton's lab can use a top down analysis to work 
backwards and determine the order of mutations in the evolution of the 
proteins that they are working with.  Behe accepts their work and likely 
hopes that such work will vindicate him, but so far it hasn't.  He 
hasn't initiated such research for any of his systems.  It is just more 
science that he doesn't want to do because he doesn't want to know the 
answer based on what has already been found.
Ron Okimoto