This is an old debate on FR. It took me a while, but I finally found Miller's (longer) online version of his book chapter, where he goes into detail about the blood clotting cascade.
Here's Behe's critique of Miller's original published description of blood clotting evolution. He complains that Doolittle's model (by way of Miller) doesn't take into account the evolution of regulatory methods, the lack of which in a modern mammal would kill it.
Here's Miller's response to Behe's critique. Here's Miller's original draft of the part of the IC chapter concerning blood clotting. As I suspected, Behe's complaints about lack of regulation destroying the organism before the new clotting steps could get refined are based on what would happen in a modern, high-blood pressure organism like humans, instead of the more realistic low blood pressure organism where the evolution of blood clotting really got started.
IOW, Behe's critique of Miller/Doolittle is an airball. It tries to cut off evolution at the start - in a small ancestor organism 600 million years ago - but does so by judging it according to the conditions of a wholly different organism!
Or as Miller puts it...
Behe asserts that the targeting of a protease, a digestive enzyme, to the bloodstream is a "potentially deadly situation," and tells the readers of his web document that we can tell how deadly this might be by looking at situations "where regulatory proteins are missing from modern organisms." In other words, Behe wants us to look at what happens when the highly-regulated current versions of clotting proteases are missing their regulatory factors. Despite this bluster, however, Behe has no evidence that the mistargeting of an inactive protease to the bloodstream would cause harm. Indeed, the recent discovery that antifreeze protein genes in fish arose from exactly such a mistargeting of proteases into the bloodstream (Chen, L., DeVries, A. L. & Cheng, C.- H. C. Proc. Natl Acad. Sci. USA 94, 38113816 (1997); and Chen, L., DeVries, A. L. & Cheng, C.-H. C. Proc. Natl Acad. Sci. USA 94, 38173822 (1997)) suggests that exactly the opposite is true.
Having made unsupported claims about the "danger" of such a mutation, Behe says that it would be difficult to see what "advantage" this would present to the organism. The answer, of course, is that it would provide a slight improvement in the organism's ability to clot blood - and that's the point. The clotting system doesn't have to work full-blast right away. In a primitive vertebrate with a low-pressure circulatory system, a very slight improvement in clotting would be advantageous, and would be favored by natural selection.
Behe then wonders how the circulating protease could become localized at the site of a clot, as if this were an insurmountable difficulty. It's not. As I suggested in my original draft on the evolution of clotting, a well-understood process called exon shuffling could have placed an "EGF domain" onto the protease sequence, and the "problem" that Behe puzzles over is solved in a flash.
Finally, Behe emphasizes that the real problem is not to generate a clot - it is to "regulate" that clot by means of an inhibitor of the protease so that it doesn't become destructive. But that's not a problem for evolution, either. As usual, Behe envisions a clotting protease that is just as powerful as the fully-evolved proteases in modern vertebrates. However, remember that this is the same guy who fretted a moment or two ago that the protease would not be strong enough to clot effectively. He wants to have it both ways. The answer to his objection is just what I wrote in the draft:
" ... a primitive clotting system, adequate for an animal with low blood pressure and minimal blood flow, doesn't have the clotting capacity to present this kind of a threat. But just as soon as the occasional clot becomes large enough to present health risks, natural selection would favor the evolution of systems to keep clot formation in check. And where would these systems come from? From pre-existing proteins, of course, duplicated and modified. The tissues of the body produce a protein known as alpha-1-antitrypsin which binds to the active site of serine proteases found in tissues and keeps them in check. So, just as soon as clotting systems became strong enough, gene duplication would have presented natural selection with a working protease inhibitor that could then evolve into antithrombin, a similar inhibitor that today blocks the action of the primary fibrinogen-cleaving protease, thrombin."
In short, none of the points raised by Behe are adequate to explain why the vertebrate clotting system could not have evolved. Furthermore, as Doolittle's work has shown clearly, the hypothesis of evolution makes testable predictions with respect to the DNA sequences of clotting proteins, and these predictions have turned out to be correct time and time again.
Why has Behe's "Biochemical Challenge to Evolution" met with so little support within the scientific community? I would suggest that the reason is simple. His hypothesis is wrong. The complex biochemical systems of living organisms, including the vertebrate clotting cascade, are fully understandable in terms of Darwinian evolution.
Here's the total response Behe had to say about that...
Kenneth Miller, Brown University Professor of Biology and author of Finding Darwin's God, has posted a response to my essays: http://biocrs.biomed.brown.edu/Darwin/DI/Design.html Overall I'm satisfied with his reply because, although he continues to defend his position, from the substance of his writing I think it should be plain to most open-minded readers that he is struggling to fend off examples that weigh heavily against Darwinism. So, for the most part, I am content to let the exchange end here. I simply urge all who are interested to read my essays as well as his response and come to their own conclusions.
Behe did respond to other points Miller has made, but I haven't followed those so I don't know if he scored points or not. But as for the blood clotting cascade, Behe clearly has pleaded no contest.