God of the Gaps...in your own knowledge. Luskin, Behe, & blood-clotting

Happy New Years and Merry Belated Kitzmas!

Well, it’s 2009 and good ol’ Casey Luskin is still fighting battles the ID advocates lost 3 years ago during the Kitzmiller case. This time the topic is the evolution of the blood-clotting cascade. Here are his 3 posts:

How Kenneth Miller Used Smoke-and-Mirrors to Misrepresent Michael Behe on the Irreducible Complexity of the Blood-Clotting Cascade: Part 1Part 2Part 3

Kenneth Miller has already posted a 3-part reply (1, 2, 3) and stolen some of my thunder (see also previous PT discussion), but I will present my own replies on some of the detailed points which Luskin missed, which happen to particularly revealing (and hilarious).


Like I said just two days after the decision came down, IDists have apparently learned nothing from defeat. They prefer to offer excuses rather than corrections to their arguments and positions. In this, history is repeating itself. After the “creation science” movement was defeated in the McLean (1981-1982) and Edwards (1981-1987) cases, creationists rained disdain and insult on the cases and courts. The judges were liberal activists, the ACLU was mean, Larry Laudan says that demarcating science from pseudoscience is hopeless, so really, judges should think that anything goes in science, including creationism, etc. They made no changes to their views, but instead made only cosmetic changes, attempting to hide the Biblical literalism and theological interventionism at the heart of their views. This, as everyone now knows, resulted in the fancy new phrase and movement “intelligent design.” But precisely because the creationists didn’t learn from their previous mistakes, and instead made excuses for their defeats, they made all the same mistakes again in the Kitzmiller case, and were defeated again. No one knows for sure what the future will hold, but I bet that what will happen is that the ID advocates will, by talking to themselves incessantly about how obviously wrong the Kitzmiller decision was on all points, completely convince themselves that Kitzmiller had no merit, and thus completely fail to learn any lessons or make any corrections. And, when/if these issues go to court again, all the same spectacular flaws will emerge again, ID (or whatever replacement term for creationism is then preferred, e.g. “critical analysis of evolution” or “strengths and weaknesses of evolution”) will again melt under expert scrutiny and cross-examination, and we’ll have yet another repeat. But we’ll see.

For the moment, the DI’s Luskin presents us with a really spectacular example of failure to learn from Kitzmiller. Luskin revisits the testimony of Ken Miller & Michael Behe on the evolution (or non-evolution) of vertebrate blood-clotting, and argues that Miller and the Court’s opinion were wrong, and Behe right. But in the process he makes a fantastic mess of the known basic science, the “irreducible complexity” argument, and the court testimony. Unraveling all of this is a fairly major project, but I will give a shot at hitting the high points.

Luskin begins by describing the situation as he sees it. Miller argued at trial that the blood-clotting cascade was not IC because whales lack one component of it, and pufferfish lack three. Behe claimed, and Luskin agrees, that Miller’s evidence didn’t undermine his argument. The judge didn’t buy Behe’s response, and Luskin doesn’t like that either:

Casey Luskin writes:

During the Kitzmiller v. Dover trial three years ago, biologist Kenneth Miller claimed that biochemist Michael Behe’s arguments in Darwin’s Black Box regarding the irreducible complexity of the blood-clotting cascade were false. Miller’s testimony led federal district court judge John Jones to assert in his decision that “scientists in peer-reviewed publications have refuted Professor Behe’s predication about the alleged irreducible complexity of the blood-clotting cascade.”

But an analysis of Miller’s arguments demonstrates that he refuted Behe in no way whatsoever, and that in fact it was Behe who refuted Miller at trial, although Judge Jones ignored Behe’s testimony. Miller continues (I am told) to go around lecturing on this topic, claiming that the blood-clotting cascade of lower vertebrates demonstrate that Behe was wrong and that the blood-clotting cascade is amenable to explanation by Darwinian evolution. Like many Darwinist claims of refutation of Behe, this one is based on smoke and mirrors.

There are minor issues here, e.g. “lower vertebrates” aren’t lower and we see once again the term “Darwinist” appearing almost as a verbal tic, but let’s get on with it and see who is really using “smoke and mirrors”:

Behe versus Pandas. (Or, Behe 1996 versus Behe 1993.)

Casey Luskin writes:

First, A Mirror

During his Kitzmiller testimony, Miller’s first misrepresentation was to equate Michael Behe’s arguments in Darwin’s Black Box regarding the blood-clotting cascade with those in the textbook Of Pandas and People (Pandas). Miller stated, “[W]hen I read through the pages of Darwin’s Black Box, I was struck by how many of the arguments used against evolution that are found in Of Pandas and People are also used in Darwin’s Black Box. And the one that really stuck in my mind was the discussion of the blood-clotting cascade in both Dr. Behe’s book and in Of Pandas and People. It struck me as essentially–the two discussions struck me as essentially identical.” (Miller, September 26 AM testimony, pg. 108, emphasis added.) So according to Miller, the treatment of the blood-clotting cascade in Pandas is “essentially identical” to the treatment of the blood-clotting cascade in Darwin’s Black Box.

The problem is that Miller’s claim is false. Behe’s treatment of the blood-clotting cascade in Darwin’s Black Box is much more precise than the treatment in Pandas, and in fact Behe made it very clear that he was limiting his argument for irreducible complexity to a particular segment of the blood-clotting cascade that had been well-studied and was well-understood.

So the “mirror” is that Miller uses Pandas’ discussion of blood-clotting in place of Behe’s allegedly more defensible argument in Darwin’s Black Box. Luskin is greatly offended at this. There are several funny things about Luskin’s position. First, in the Kitzmiller case, Of _Pandas and People_ (2nd edition, 1993) was the primary issue, NOT BEHE. It was Pandas that was recommended in the Dover school board’s ID policy, to give students an understanding “of what intelligent design actually involves”, not anything by Behe. It was 58 or so copies of Pandas that were bought with church offerings and forced into Dover classrooms or libraries, not anything by Behe. It was Pandas which was cited in the plaintiffs’ initial complaint to the Court, and therefore, rationally and logically and completely common-sensically, it was Pandas that was the primary target of the plaintiffs’ expert witnesses. If Darwin’s Black Box had been the recommended text, things might have been (a little) different, but it wasn’t. Failure to acknowledge this basic point is a colossal flaw in Luskin’s whole argument.

But what is really, really, really funny here is a basic fact, now well-known (or at least it should be), which Luskin appears to be completely unaware of. Luskin takes great pains to distinguish Behe’s treatment of blood-clotting from the apparently less-defensible treatment in Pandas, and expresses great indignation at Kenneth Miller for conflating the two. How dare the great Behe be mixed in with that old half-baked textbook!, Luskin apparently thinks. But what Luskin doesn’t get, astonishingly and hilariously, is that Behe himself wrote the blood-clotting section of the 1993 edition of Pandas_!_

It is true that this wasn’t well-known in early 2005. In fact, it was essentially unknown in 2005. Behe is not listed as an author on the 1993; instead, he is just one of dozens of reviewers (most of them known young-earth or old-earth creationists, by the way) listed in the front matter of the book. The listed authors on the 1st (1989) and 2nd (1993) editions are Dean Kenyon (young-earth creationist) and Percival Davis (young-earth creationist who is still teaching flood geology at a Bible college to this day!) [1]. Behe is a new addition to the reviewers list between the 1989 and 1993 versions, but there is no indication that he played any role beyond that of other reviewers.

I only realized that there was more to it in about March 2005 when, during various Kitzmiller-research activities, I accidentally read Behe’s blood-clotting chapter in Darwin’s Black Box near to the time that I read the blood-clotting discussion in Pandas. I found this sort of eerie similarity:

“When a container of liquid, like a can of soda, springs a leak, the fluid quickly drains out. However, when a person suffers a cut it ordinarily bleeds for only a short time before a clot forms to stop the bleeding. Soon the clot hardens and eventually the cut heals over. Blood clot formation seems so familiar to us that most people don’t give it much thought. However, biochemical investigation has shown that blood clotting is a very complex, intricately-woven system containing a score of interdependent protein parts. The absence or defective operation of any of several of these components will cause the system to fail, and blood will not clot at the proper time or at the proper place.” (Pandas 1993, p. 141)

“When a container of liquid – like a carton of milk, or a tank truck filled with gasoline – springs a leak, the fluid drains out. The rate of flow can depend on the thickness of the liquid (for example, maple syrup will leak more slowly than alcohol), but eventually it all comes out. No active process resists it. In contrast, when a person suffers a cut it ordinarily bleeds for only a short time before a clot stops the flow; the clot eventually hardens, and the cut heals over. Blood clot formation seems so familiar to us that most people don’t give it much thought. Biochemical investigation, however, has shown that blood clotting is a very complex, intricately woven system consisting of a score of interdependent protein parts. The absence of, or significant defects in, any one of a number of the components causes the system to fail: blood does not clot at the proper time or at the proper place.” (Behe, Darwin’s Black Box, 1996, p. 78)

Whoa! Look at that. The obvious interpretation was that Behe anonymously authored a chunk of the 1993 Pandas, presumably the whole new chunk from pages 141-146 on blood-clotting and proteins (which was all new in the 2nd edition).

In the entirety of the literature on or about the ID movement this very interesting fact about Behe’s involvement in Pandas is never mentioned, except one brief, vague reference in Thomas Woodward’s 2003 cheer-leading history of ID, Doubts about Darwin, where he writes,

“Michael Behe assisted in the rewriting of a chapter on biochemistry in a revised edition of Pandas. The book stands as one of the milestones in the infancy of Design.” (Woodward, 2003, p. 89)

The only other pre-_Kitzmiller_ reference I ever found to Behe authoring Pandas was a long-forgotten 1994 hand-written note by Genie Scott in the NCSE archives. Scott and Behe had participated in some sort of encounter on CNN, when Behe was a new and probably mostly unheard of figure in the antievolution movement, and apparently during the show, or the prep for the show, it was mentioned that Behe was an author of Pandas, which Scott jotted down amongst other notes.

Anyway, the textual evidence, plus Woodward’s statement, was convincing, but just to be sure, we asked Behe about this during his May 2005 sworn deposition. He confirmed that he authored pages 141-146 of Pandas, although he seemed at a loss to explain why he was listed as a “reviewer” rather than author. Behe also confirmed his authorship at trial at various points in his testimony. It has since been widely mentioned, e.g. in my articles, in Ken Miller’s new book, etc., but it still surprises people, primarily those who mostly heard of ID through Johnson and Darwin on Trial, and Behe and Darwin’s Black Box, and assume, incorrectly, that knowing about the most famous ID works and authors provides an accurate first approximation of the history of the ID movement.

But it is quite surprising that Luskin, of all people, isn’t aware of it. The only suggestion I can think of is that he enjoys reading the work of ID critics who are not very serious, but finds reading the works of the people who really know their stuff and have really got ID’s number…say, I don’t know, Kenneth Miller and PT authors…too painful.

One aside: the fact that Behe wrote a chunk of Pandas is important in several ways apart from pure history. First, this makes Pandas, rather than Darwin’s Black Box (or really, a few of his web articles), the first published expression of Behe’s IC argument. Second, it means that Behe, like all of the other major players in the ID movement, pretty clearly endorsed the ID movement’s get-into-the-public-schools-first, do-the-scientific-research-later philosophy and practice. Third, it nukes Luskin’s indignancy about Miller failing to distinguish the blood-clotting treatments in Darwin’s Black Box and Pandas; these are really just two closely related versions of the same argument, by the same person, Behe 1993 and Behe 1996. It gets even worse for Luskin once you realize that Behe defended both treatments as correct in his Kitzmiller testimony, as we will discuss below.

How many parts in the irreducible system, really?

So, we’re done with the mirror. Let’s continue to the smoke:

Casey Luskin writes:

Then A Puff of Smoke

By equating Behe’s treatment of blood clotting with that of Pandas (see above), and by quoting Pandas’ statement that “Only when all the components of the [blood clotting] system are present and in good working order does the system function properly,” Miller implied to Judge Jones that according to Darwin’s Black Box the entire blood-clotting cascade is irreducibly complex. Wrong. While Pandas made the claim of irreducible complexity with respect to the entire blood-clotting cascade, Behe in Darwin’s Black Box did not.

Take note: Luskin has just admitted that Behe contradicted himself in his definition of the “irreducible” blood-clotting system, in two accounts between 1993 and 1996. Since Behe defended both accounts in the Kitzmiller trial in 2005, and Luskin is defending Behe in 2008, some combination of Behe 2005, Luskin 2008, Behe 1996, and Behe 1993 is wrong about something. It doesn’t really matter which, because whichever ones you pick, Luskin’s attack on Miller and the Kitzmiller opinion is toast.

Luskin continues:

Casey Luskin writes:

To understand the difference between Behe’s views and the account in Pandas,

…which were also Behe’s views…hilarious!

one needs to understand some basics about blood-clotting cascades.

This ought to be good…

Casey Luskin writes:

Roughly speaking, in land-dwelling vertebrates, there are two different pathways by which the blood-clotting cascade can be initiated – the “intrinsic” pathway, and the “extrinsic” pathway. (There can be some crossover between the two pathways.) The final stages of the blood-clotting cascade take place after either pathway reaches factor X, also called the Stuart factor. These final stages of the cascade are what Behe calls “beyond the fork” or “after the fork.” Figure 1 below contains a very rough description of how the blood-clotting cascade can be initiated by either the extrinsic or intrinsic pathway before it forms a final clot, and Figure 2 contains a full description of the land-dwelling vertebrate blood-clotting cascade.

Luskin uses the awkward phrase “land-dwelling vertebrate” because he apparently knows only 3 things: (1) there is something called the “vertebrate blood-clotting system”, (2) Ken Miller pointed out that whales/dolphins are missing a piece, and (3) Ken Miller pointed out that fishes are missing several pieces. So, like confidently ignorant creationists everywhere, Luskin naively and blissfully assumes that the little bit of knowledge that has been forced upon him by evolution educators represents the sum total of relevant knowledge, and that this knowledge represents reality. I.e., he assumes that “land-dwelling vertebrates” must have the canonical system found in humans and most mammals and diagrammed in Pandas and Darwin’s Black Box. But, as anyone acquainted with evolution would expect, the “canonical” well-studied system actually has many known variants even with the limited knowledge now available:

FXII is absent from fish; it is present in frog, platypus, and opossum, but is absent in chicken, an apparent example of gene loss. A single gene corresponding to the evolutionary predecessor of FXI and prekallikrein occurs in frog, chicken, and platypus. (Ponczek MB, Gailani D, Doolittle RF. (2008). “Evolution of the contact phase of vertebrate blood coagulation.” J Thromb Haemost. 2008 Nov;6(11):1876-83.

Ooops. Well, anyway, what does Luskin think is Behe’s One True Definition of the Really-Irreducible-For-Real-This-Time blood clotting cascade?

Casey Luskin writes:

In Darwin’s Black Box, Behe specifically stated that his argument for irreducible complexity only pertained to irreducible complexity “beyond the fork” where the intrinsic and extrinsic blood-clotting cascades converge. As Behe writes:

Leaving aside the system before the fork in the pathway, where some details are less well known, the blood-clotting system fits the definition of irreducible complexity. … The components of the system (beyond the fork in the pathway) are fibrinogen, prothrombin, Stuart factor, and proaccelerin. … in the absence of any one of the components, blood does not clot, and the system fails. (Michael Behe, Darwin’s Black Box: The Biochemical Challenge to Evolution, pg. 86 (Free Press, 1996), emphasis added.)

Behe even explained this very point to Judge Jones, making it clear that his own argument was not as expansive as that of Pandas:

The relative importance of the two pathways in living organisms is still rather murky. Many experiments on blood clotting are hard to do. And I go on to explain why they must be murky. And then I continue on the next slide. Because of that uncertainty, I said, let’s, leaving aside the system before the fork in the pathway, where some details are less well-known, the blood clotting system fits the definition of irreducible complexity. And I noted that the components of the system beyond the fork in the pathway are fibrinogen, prothrombin, Stuart factor, and proaccelerin. So I was focusing on a particular part of the pathway, as I tried to make clear in Darwin’s Black Box. If we could go to the next slide. Those components that I was focusing on are down here at the lower parts of the pathway. And I also circled here, for illustration, the extrinsic pathway. It turns out that the pathway can be activated by either one of two directions. And so I concentrated on the parts that were close to the common point after the fork. So if you could, I think, advance one slide. If you concentrate on those components, a number of those components are ones which have been experimentally knocked out such as fibrinogen, prothrombin, and tissue factor. And if we go to the next slide, I have red arrows pointing to those components. And you see that they all fall in the area of the blood clotting cascade that I was specifically restricting my arguments to. And if you knock out those components, in fact, the blood clotting cascade is broken. So my discussion of irreducible complexity was, I tried to be precise, and my argument, my argument is experimentally supported.

(Michael Behe, Oct. 18 testimony, pg. 25-28, emphasis added)


Behe provided Judge Jones with an experimentally verified case for irreducible complexity of the blood-clotting cascade with respect to particular proteins within the cascade: fibrinogen, prothrombin, and tissue factor. He discussed the cascade “after the fork” where the intrinsic and extrinsic pathways converge. Behe didn’t want to extend his argument too far, because adequate experimental tests had not yet been performed to demonstrate irreducible complexity with respect to some of the other factors, particularly those involved in the intrinsic pathway.

There’s Luskin’s key message. Behe really meant that just a small portion of the blood-clotting system was IC, not the whole thing! Surprise! There are several problems with this. First, as Kenneth Miller has pointed out, the reader does not even get this message consistently from Behe’s (1996) Darwin’s Black Box. He cites several places where Behe says that the whole system (and every part!) is IC. In the interests of pedantry, I did a survey of Behe 1996:

Over the next few pages you will meet the score of protein players in the game of blood clotting and learn a bit about their roles. (Behe 1996, p. 79)

A score of different protein players? That’s a lot more impressive-sounding than four parts, which is what Luskin insists is Behe’s Really and For Real definition of blood-clotting. But, being fair, Behe is just starting his discussion. Behe spends several pages tracing the blood-clotting cascade backwards from the final result (which is fibrin, the main clotting protein produced by the cutting of fibrinogen). At the end of this discussion, Behe discusses the intrinsic and extrinsic pathways, which both activate Stuart factor and therefore rest of the pathway. He notes some ambguity about how distinct these pathways are:

There is also an important conceptual similarity between the Foghorn attack system and the blood-clotting pathway: both are irreducibly complex. Leaving aside the system before the fork in the pathway, where some details are less well known, the blood-clotting system fits the definition of irreducible complexity. That is, it is a single system composed of several interacting parts that contribute to the basic function, and where the removal of any one of the parts causes the system effectively to cease functioning. The function of the blood clotting system is to form a solid barrier at the right time and place that is able to stop blood flow out of an injured vessel. The components of the system (beyond the fork in the pathway) are fibrinogen, prothrombin, Stuart factor, and proaccelerin. Just as none of the parts of the Foghorn system is used for anything except controlling the fall of the telephone pole, so none of the cascade proteins are used for anything except controlling [he formation of a blood clot. Yet in the absence of any one of the components, blood does not clot, and the system fails. (Behe 1996, pp. 85-86)

This is the passage upon which Luskin (and Behe in 2005) lay great emphasis, and great indignation towards Kenneth Miller.

But Behe’s discussion of the IC-ness of the blood-clotting cascade doesn’t stop on page 86 of Darwin’s Black Box. Behe goes on to wonder if a simpler blood-clotting system is possible:

One could imagine a blood-clotting system that was somewhat simpler than the real one – where, say, Stuart factor, after activation by the rest of the cascade, directly cuts fibrinogen to form fibrin, bypassing thrombin. Leaving aside for the moment issues of control and timing of clot formation, upon reflection we can quickly see that even such a slightly simplified system cannot change gradually into the more complex, intact system. If a new protein were inserted into the thrombin-less system it would either turn the system on immediately – resulting in rapid death – or it would do nothing, and so have no reason to be selected. Because of the nature of a cascade, a new protein would immediately have to be regulated. From the beginning, a new step in the cascade would require both a proenzyme and also an activating enzyme to switch on the proenzyme at the correct time and place. Since each step necessarily requires several parts, not only is the entire blood-clotting system irreducibly complex, but so is each step in the pathway. (pp. 86-87, bold added)

Whoops! Which Behe has the One True Definition? Behe page 86, or page 87?

I think a ship canal is a good analogy for this aspect of the blood-clotting system. The Panama Canal allows ships to cross the Isthmus from the Pacific Ocean to the Caribbean Sea. Because the land is higher than sea level, water in a lock lifts a ship up to a level where it can travel along for a while. Then another lock lifts the ship to the next level, and locks on the other side lower the ship back down to sea level. At each lock there is a gate that holds back the water as the ship is raised or lowered; there is also a sluice or water pump that drains or fills the lock. From the beginning each lock must have both features – a gate and a sluice – or it does not function. Consequently, each of the locks along the canal is irreducibly complex. Analogously, each of the control points of the blood-clotting cascade needs both an inactive proenzyme and a separate enzyme to activate it.

He goes on to compare “this aspect of the blood-clotting system” (the regulation cascade) to the series of locks on the Panama canal:

From the beginning each lock must have both features – a gate and a sluice – or it does not function. Consequently, each of the locks along the canal is irreducibly complex. Analogously, each of the control points of the blood-clotting cascade needs both an inactive proenzyme and a separate enzyme to activate it. (Behe, p. 87)

Whoops! Each of the control points is now IC. That’s 2 votes for Miller’s interpretation, and 1 for Behe/Luskin’s.

Behe then goes on to discuss the processes that halt clotting (pp. 87-88) in a section heading entitled “IT’S NOT OVER YET.” Behe wraps up:

The formation, limitation, strengthening, and removal of a blood clot is an integrated biological system, and problems with single components can cause the system to fail. The lack of some blood clotting factors, or the production of defective factors, often results in serious health problems or death. The most common form of hemophilia arises from a deficiency of antihemophilic factor, which helps activated Christmas factor in the conversion of Stuart factor to its active form. Lack of Christmas factor is the second most common form of hemophilia. Severe health problems can also result if other proteins of the clotting pathway are defective, although these are less common. Bleeding disorders also accompany deficiencies in FSF, vitamin K, or [alpha]2-antiplasmin, which are not involved directly in clotting. Additionally, lack of protein C causes death in infancy due to the occurrence of numerous, inappropriate clots. (Behe, pp. 88-89, bolds added)

The bolded passage raises the vote to 4-1 for Miller’s interpretation. Plus, Behe lists six other proteins or cofactors, none of them in his minimal 4-protein system, as examples that “cause the system to fail.”

Various additional examples pile up:

Is it possible that this ultra-complex system could have evolved according to Darwinian theory? (p. 89)

Is Behe really just talking about the 4-protein mini-system as “ultra-complex”?

Doolittle’s scheme runs into the same problem in the production of prothrombin, a thrombin receptor, antithrombin, plasminogen, antiplasmin, proaccelerin, Stuart factor, proconvertin, Christmas factor, antihemophilic factor, and protein C – virtually every protein of the system! (p. 94)


Since two proteins – the proenzyme and its activator – are both required for one step in the pathway, then the odds of getting both the proteins together are roughly the square of the odds of getting one protein. […] Doolittle’s scenario implicitly acknowledges that the clotting cascade is irreducibly complex, but it tries to paper over the dilemma with a hail of metaphorical references to yin and yang. The bottom line is that clusters of proteins have to be inserted all at once into the cascade. This can be done only by postulating a “hopeful monster” who luckily gets all of the proteins at once, or by the guidance of an intelligent agent. (p. 96)

Blood coagulation is a paradigm of the staggering complexity that underlies even apparently simple bodily processes. Faced with such complexity beneath even simple phenomena, Darwinian theory falls silent. (p. 97)

Like some ultimate Rube Goldberg machine, the clotting cascade is a breathtaking balancing act in which a menagerie of biochemicals – sporting various decorations and rearrangements conferred by modifying enzymes – bounce off one another at precise angles in a meticulously ordered sequence until, at the denouement, Foghorn Leghorn pushes off the telephone pole and gets up from the ground, the bleeding from his wounds stopped. The audience rises to its feet in sustained applause. (p. 97)

At none of these points later in the chapter does Behe suggest or imply he is talking about a mere 4-protein system. The vote is about 8-1. For confirmation, we can look elsewhere in the book. For example, in Behe’s wrap-up chapter, where he summarizes each system:

The function of the blood-clotting system is as a strong, but transient barrier, The components of the system are ordered to that end. Fibrinogen, plasminogen, thrombin, protein C, Christmas factor, and the other components of the pathway together do something that none of the components can do alone. When vitamin K is unavailable or antihemophilic factor is missing, the system crashes just as surely as a Rube Goldberg machine fails if a component is missing.The components cut each other in precise places, align with each other in exact ways. They act to form an elegant structure that accomplishes a specific task. (p. 204, bold added)

So now the vote is 9-1 (he actually emphasizes the ICness of the larger system twice here, but I’m being generous).

A Rube Goldberg machine is like the blood-clotting system in that they are both irreducibly complex, even though they have many differences. In order to reach a conclusion based on an analogy, it is only necessary that the deduction flow out of the shared properties: The irreducibly complex Rube Goldberg machine required an intelligent designer to produce it; therefore the irreducibly complex blood-clotting system required a designer also. (p. 218)

Nothing about only 4 proteins, and any decent Rube Goldberg device has more than 4 parts! 10-1.

As if that isn’t enough, we can look to other articles by Behe. For example, here’s a 1996 speech Behe gave at the Discovery Institute:

When an animal is cut a protein called Hageman factor sticks to the surface of cells near the wound. Bound Hageman factor is then cleaved by a protein called HMK to yield activated Hageman factor. Immediately the activated Hageman factor converts another protein, called prekallikrein, to its active form, kallikrein. Kallikrein helps HMK speed up the conversion of more Hageman factor to its active form. Activated Hageman factor and HMK then together transform another protein, called PTA, to its active form. Activated PTA in turn, together with the activated form of another protein (discussed below) called convertin, switch a protein called Christmas factor to its active form. Activated Christmas factor, together with antihemophilic factor (which is itself activated by thrombin in a manner similar to that of proaccelerin) changes Stuart factor to its active form. Stuart factor, working with accelerin, converts prothrombin to thrombin. Finally thrombin cuts fibrinogen to give fibrin, which aggregates with other fibrin molecules to form the meshwork clot you saw in the last picture.

Blood clotting requires extreme precision. When a pressurized blood circulation system is punctured, a clot must form quickly or the animal will bleed to death. On the other hand, if blood congeals at the wrong time or place, then the clot may block circulation as it does in heart attacks and strokes. Furthermore, a clot has to stop bleeding all along the length of the cut, sealing it completely. Yet blood clotting must be confined to the cut or the entire blood system of the animal might solidify, killing it. Consequently, clotting requires this enormously complex system so that the clot forms only when and only where it is required. Blood clotting is the ultimate Rube Goldberg machine. (Michael J. Behe, “Evidence for Intelligent Design from Biochemistry.” From a speech delivered at Discovery Institute’s God & Culture Conference, Discovery Institute, August 10, 1996)

11-1. Or in a reply to Shanks and Joplin in 2000:

Yet contrast this case with that of mice in which the gene for either fibrinogen (Bugge et al. 1996a), tissue factor (Bugge et al. 1996b), or prothrombin (Sun et al. 1998) has been knocked out. Those proteins are all components of the blood clotting cascade, which I discussed prominently in Darwin’s Black Box (1996, Ch. 4), claiming it is irreducibly complex. The loss of any one of those proteins prevents clot formation–the clotting cascade is broken. Thus Shanks and Joplin’s concept of redundant complexity does not apply to all biochemical systems.

Tissue factor is not one of the 4 components, and Behe says nothing about a minimal 4-protein system here. 12-1.

If anyone was really interested, we could go on and survey what all of the other ID proponents have said about the definition of the blood-clotting cascade. I have not done this systematically [2], but my sense of it is that they are always referring to the full system, and the alleged 4-protein “really and truly IC!” system is never mentioned until Behe made the argument in response to Ken Miller in the 2005 Dover trial.

But what does Pandas say?

I almost forgot another major problem with Behe’s claim that the “real” thing he meant to call IC was the 4-protein subsystem consisting of “fibrinogen, prothrombin, Stuart factor, and proaccelerin.” As it turns out, Behe’s treatment in Pandas says something different. But wait, Luskin thinks that Behe’s 1996 is different from the flawed treatment in Pandas. That’s nice, but does Behe agree?

Here is what Behe said during his direct testimony, that is, while he was being questioned by his pro-ID lawyer:

Q. Now you referred to, I believe, a blood clotting cascade, is that correct?

A. That’s right.

Q. Can you explain a little bit to us as you’re explaining what we see here on this particular diagram?

A. Okay, sure. Yeah, this is a figure of the blood clotting cascade taken from the biochemistry textbook by Voet and Voet, which is widely used in colleges and universities around the country. You see all these names of things and arrows. The names of things are very complex proteins of the complexity or sometimes more complex than the hemoglobin that I showed yesterday.

In blood clotting, the material that forms the clot cannot, of course, be in its solid clotted form during the normal – during the normal life of an animal or all of the blood would be clotted, and that would be inconsistent with its life. So the material of the clot that actual eventually forms the clot exists as something called fibrinogen, which is actually a soluble pre-cursor to the clot material.

It floats around in your bloodstream during normal times. But when a cut occurs, fibrinogen is transformed into something called fibrin, and that happens when another protein comes along and cuts off a small piece of fibrinogen, a specific piece which exposes a sticky site on it, sticky in the sense of those two proteins yesterday that I saw that – that I showed you that had complimentary surfaces.

It exposes a sticky site on the surface of the fibrinogen, which allows the many copies of fibrinogen, now turned into fibrin, to aggregate and stick to each other, forming the blood clot.

But what is the component that cuts fibrinogen and activates it? Well, the component is another protein called thrombin. But now we’ve got the same problem again. If thrombin were going around cutting fibrinogen and turning it into fibrin, all the blood would clot, and that would congeal the blood and kill the animal.

So thrombin itself is an inactive form called prothrombin, so it has to be activated when a cut occurs. And that’s the responsibility of another protein. And that protein exists in an inactive form, and it’s – the activation of that is the responsibility of another protein.

So in the blood – it’s called a blood clotting cascade because one component acts on the next which acts on the next which acts on the next and so on. Now notice that the blood clotting cascade actually has what are called two branches. There is one in this box up here is labeled the intrinsic pathway. And this is labeled the extrinsic pathway. So there are actually two branches to this blood clotting cascade.

Q. I believe this section is addressed in the textbook Pandas, correct?

A. Yeah, that’s correct. On the left is a figure from Of Pandas and People illustrating the blood clotting cascade. And that was drawn after the illustration from the textbook by Voet and Voet. On the right-hand side is the illustration for the blood clotting cascade that appears in Darwin’s Black Box.

I discussed the blood clotting cascade in one chapter of that – of my book, and the illustration is very similar to the one in Pandas.

Q. I believe the diagram in Pandas is found on page 143?

A. Yes, that’s right.

Q. Now these two diagrams, the one that appears in Darwin’s Black Box and one of the blood clotting cascade appear, to my eye, to be virtually similar or almost exactly similar?

A. Yeah, they are very similar, except for the color in Pandas and so on. And that’s because I wrote the discussion in Pandas and, of course, also in my own book. So the figures are very similar between the two.

Q. Now you testified yesterday that you coined the term irreducible complexity in Darwin’s Black Box, which was published in 1996, is that correct?

A. Yes.

Q. So that book was published actually three years after Pandas was written, is that accurate?

A. Yes, that’s correct.

Q. Is it accurate to say then that the concept of irreducible complexity was not fully developed when you had written that section in Pandas on blood clotting in 1993?

A. Yes, that’s right. I was still contemplating the idea.

Q. Does Pandas, however, discuss the complexity of this system, the blood clotting system?

A. Yes, it does. It elucidates all the parts of the system.

Q. Is that discussion consistent with your discussion in Darwin’s Black Box?

A. Yes, it introduces the concept of the purposeful arrangement of parts and says that’s how we perceive design.

Q. That’s introduced in the Pandas book?

A. Yes, uh-huh.

Q. When you talk about the purposeful arrangement of parts, that’s similar to what you were discussing yesterday in your testimony, is that correct?

A. Yes.

**Q. So is the scientific explanation of the blood clotting system similar to the – the discussion in Pandas similar to the blood clotting cascade scientific explanation in Darwin’s Black Box?

A. That’s right, they’re essentially the same. I think it’s more detailed in Darwin’s Black Box.** (Behe’s Kitzmiller testimony, Day 2, online here. Bold added.)

Whoops! “Essentially the same” was Behe’s initial description of the blood-clotting discussions in Pandas and Darwin’s Black Box. In his direct testimony he even pointed out that basically the same diagrams were used – something Kenneth Miller also pointed out.

This is not good for Luskin, who takes great pains to distinguish the two treatments, to keep the venerable Behe’s authority away from the troublesome problems with Pandas.

Of course, in his direct testimony, Behe goes on to rebut Ken Miller’s testimony, relying on the “I just meant the 4 proteins” defense. Luskin puts great weight on Behe’s response.

I, and Ken Miller, have already discussed the problems with this just based on a close reading of Darwin’s Black Box. But what, I wonder, does Behe’s “essentially the same” treatment in Pandas say about those same four proteins?

We may try many smaller sets of components to get started – fibrinogen, prothrombin, activated Stuart factor and proaccelerin, or inactive Stuart factor or proaccelerin, or fibrinogen plus an imaginary protein that cleaves fibrinogen to fibrin – death is nearly always the certain result. (Pandas 1993, page 145)

This, as they say, is a Big Oops. In his 2005 direct testimony, Behe belabored how he had been victimized by Ken Miller ignoring his 4-protein definition. On the 4-protein definition, all of Miller’s missing-proteins evidence was irrelevant, said Behe. Luskin accepts this whole-heartedly and bashes the Kitzmiller Court for “ignoring” it. But who is really ignoring testimony here? Luskin only mentioned Behe’s direct testimony, but ignored a rather important bit…the famous cross-examination by Eric Rothschild.

I’m really surprised Luskin didn’t think to mention this, considering that he watched the Behe testimony (and cross, I thought, but I don’t specifically remember), and I’ve already written up this episode of the Rothschild-Behe battle for NCSE Reports. I’ll just quote that:

Unbeknown to Behe, Eric Rothschild had been plotting his cross-examination for months, with help from Kenneth Miller, Kevin Padian, me, and numerous others, including random members of the public offering unsolicited e-mail suggestions (“When you get Behe on the stand, you have to ask him this …”). Rothschild had assembled several dozen lines of questioning that would dissect the irreducible complexity argument, its various supporting examples, and perhaps more importantly the indignant rhetoric that Behe uses to give the impression of an impressive scientific case. Rothschild showed numerous contradictions between Behe’s statements and the published evidence (for example, the immune system episode), and between different statements made by Behe. A particularly impressive example of the latter involved blood-clotting.

Rothschild noticed something that I had not in the 1993 edition of Pandas: in 1993, Behe (who wrote the blood-clotting section on p 141-6 of the 1993 Pandas, although he is not listed as an author) defined the irreducibly complex blood-clotting system differently than he did in his 1996 Darwin’s Black Box. In 1993, Behe said that if an organism had only the four core components of the blood-clotting system (“Stuart factor and his friends,” as Rothschild put it), it would have a nonfunctioning system and would die. However, in 1996, Behe, presumably having become aware of the fact that there is a fair bit of variability in vertebrate blood-clotting systems, said that those four proteins constituted “the system”, and furthermore at trial, Behe criticized Ken Miller for ignoring this definition. Rothschild, with mock innocence and a big grin, pointed out the discrepancy, and then let Behe attempt to invent a rationale on the fly. Behe ended up coming up with yet another definition of “the system”, but the point was made – Behe protected his irreducible complexity argument from what would otherwise be a clear falsification by redefining the “irreducibly complex system” at will. Contradictory evidence was dodged with word games, rather than accepted. Rothschild set up example after example of this sort of thing, and each time Behe would exercise his substantial powers of rationalization to paper over the problem, or define it away, or provide some excuse about why evolution had produced the scientific goods and ID had not. (Nick Matzke, 2006, “Design on Trial: How NCSE Helped Win the Kitzmiller Case.” 26(1-2), pp. 37-44.)

This portion of the cross-examination is here. If you read it, you can see Rothschild simply draws out the discrepancy and asks Behe to explain it. Behe could have just said “I was wrong in Pandas, my newer definition is right.” But of course, the whole point of Behe being there was to defend the ID book on trial, which was Pandas, so he couldn’t do that. He also couldn’t say Pandas was right, and his newly-minted 4-proteins defense against Ken Miller was wrong because then his main anti-Miller point on blood-clotting was shot. He was in a pickle, tried to rationalize his way out of it, and it was apparent to the judge and most of the audience (except, perhaps, Luksin). Rothschild lined up dozens of these sorts of things, and the effect on Behe’s credibility was devastating. Judge Jones has since confirmed in interviews that this was his impression upon watching the cross as well.

In summary, not only is Behe’s only-4-proteins defense dubious on its own terms, it was explicitly contested at trial during cross-examination. The judge heard both sides and agreed with one of them, and put it in the decision. This is how things go. Luskin and ID advocates are free to disagree with the judge, but they have no hope of convincing reasonable observers if they ignore key relevant events in the court record (the sworn testimony, which is the most important evidence a judge relies on), as Luskin did here.

Other stuff: Bicycles and the definition of “function”

Since I’m on a roll, I will highlight a few other weird features of Luskin’s treatment of Behe & blood-clotting.

Casey Luskin writes:

The Blood-Clotting Cascade’s Irreducible Core

The blood-clotting cascade pathway of jawed fish has an important difference from that of land-dwelling vertebrates because fish don’t have an intrinsic pathway found in land-dwelling vertebrates. That doesn’t mean that the rest of the cascade isn’t irreducibly complex, for both may have a core system of parts that is irreducibly complex. And how do we know there’s an irreducible core comprised of parts Miller didn’t address in his testimony? Because experimental evidence, cited by Michael Behe during the Kitzmiller trial and in Darwin’s Black Box, shows that some components of this system are absolutely necessary to have a functional blood-clotting system.

The “irreducible core” is a long-standing concept within ID thinking that William Dembski and Jonathan Wells define as follows:

A functional system is irreducibly complex if it contains a multipart subsystem (i.e. a set of two or more interrelated parts) that cannot be simplified without destroying the system’s basic function. We call this multipart subsystem the system’s irreducible core. … We therefore define the core of a functionally integrated system as those parts that are indispensible to the system’s basic function: remove parts of the core, and you can’t recover the system’s basic function from the other remaining parts.

(Jonathan Wells and William Dembski, The Design of Life: Discovering Signs of Intelligence in Biological Systems, pgs. 146-147 (Foundation for Thought and Ethics, 2008).)

Dembski also discussed the concept of the “irreducible core” in his 2001 book No Free Lunch where he wrote, “Consider an old-fashioned pocket-watch with a winding mechanism. The basic function of the watch is to tell time. What’s more, many parts of the watch are indispensible to that basic function, for instances, the spring, the face, the hand, and the minute hand. On the other hand, other parts of the watch are dispensable, for instance the crystal, the metal cover holding the crystal, and the chain. By focusing purely on the indispensible parts of the pocket watch one obtains what can be called an irreducible core that has all the crucial properties of irreducibly complex systems considered so far. It therefore makes sense to define an irreducibly complex system as one that contains an irreducible core whose parts are each indispensible, but where the system is itself permitted to certain unnecessary or redundant elements.” (No Free Lunch, pg. 285)

Here Luskin cites people other than Behe, writing after Behe, and making an argument which Behe didn’t make at trial. The “irreducible core” idea was invented as a defensive measure once it began to become clear that a lot (all of them, actually! I’ve kept track!) of Behe’s “irreducible” systems were actually more reducible than ID advocates thought. Basically it is a hidden admission that the original argument was flawed. The originally-identified system is shown to be reducible, and so the “irreducibility” claim is pulled back in a cowardly effort to avoid admitting the obvious point, which is that if each of these systems has numerous parts which are required in some organisms but not in the same systems with the same function in other organisms, then Behe’s argument that these systems couldn’t evolve because partial systems would be “by definition nonfunctional” (Behe 1996, p. 39) is shot.

Then we get to the really funny stuff:

Casey Luskin writes:

I like to explain the “irreducible core” using the analogy of a bicycle: A bicycle has an irreducible core that requires a frame, two wheels, a motor mechanism (like legs on pedals), and a steering mechanism (like handle-bars attached to the front wheel). A bicycle also has a seat, but obviously you can ride a bike without a seat (though it wouldn’t be very fun). So, while the seat sure helps a lot, it is not part of the irreducible core of a bike. Same could be said for light deflectors, etc. So the fact that a bike has a couple dispensable parts doesn’t mean that there isn’t an irreducible core to a bike.

When I first read this, I actually though for a bit that Luskin might have made a little bit of a point. Depending on how you define “function”, I suppose you could say that a bicycle is irreducible despite the existence of a unicycle. If the function is “two-wheeled tranport” then, sure, two wheels are required and the system is “irreducible”.

Of course this relies on making the definition of “function” horribly subjective and therefore unscientific and useless. You could make anything “IC” by including the details of the parts in the definition of the “function.” For example, you might not think that a hood ornament is an irreducible part of a car, but it is irreducible if the function is “driving with a hood ornament.” This sort of silliness is only invoked by craven ID advocates who are afraid of admitting and correcting the errors of their Great Authority, Behe. The right definition of “function” in these arguments, of course, was originally in reference to natural selection. That’s how Behe put it in 1996, and that’s how the argument is supposed to work, because partial systems weren’t supposed to be selectable, and therefore allegedly natural selection couldn’t produce them. [3]

Anyway, all of this is academic, since Carl Zimmer found out that, in fact, removing a wheel is not fatal to the function of a bicycle, which is rolling motion:

Whoops! It doesn’t even look that hard!

There is one other issue with Luskin’s bicycle/unicycle analog. Bicycles and unicycles, somewhat arguably, work on different principles (not that different, but whatever). This is not the case with the different blood-clotting systems. They are basically the same components, mostly with homologous proteins serving the same role. Quite probably many of the proteins would be interchangable between organisms. The components that are “missing” in dolphins/whales and fishes are, I believe, just part of the signal-amplification part of the cascade. An organism with more of them might have a more rapid or stronger or sensitive response to an injury. These sorts of proteins could be favored by selection initially just to improve clotting, and then become effectively essential in some cases due to some combination of secondary changes to the cascade (other parts of the case coming to rely on what was just helpful initially) or changed organismal circumstances (e.g., active organisms with high blood pressure may need a more sensitive system). Variability in the presence and required-ness of these sorts of proteins directly supports this hypothesis.

So if we are going to insist on analogies (which in my experience confuses arguments about IC as much as it clarifies them, since then pointless arguments about “evolving” bicycles or mousetraps begin to occur), the difference between the various blood-clotting cascades which Miller cited is more like the difference between bicycles with different numbers of gears. To get a new gear, one could just duplicate an old gear, and then modify it slightly. This, as it happens, is basically how many of the “irreducible” parts of the full-blood clotting system were added.

Moving on, here is Luskin invoking the what-we-thought-was-IC-isn’t-so-this-subset-MUST-be-for-sure-irreducible argument:

Casey Luskin writes:

By finding that both jawed vertebrates and land-dwelling vertebrates contain all of the above components – the extrinsic pathway and everything “after the fork” – Miller’s arguments, at most, show that there is an irreducible core comprised of the components in the above diagram. Perhaps the intrinsic pathway of land-dwelling vertebrates (missing from this diagram) is not part of this irreducible core. If that’s the case, this does not refute irreducible complexity for the rest of the system; it just shows that those components in the intrinsic pathway aren’t indispensable to the system. The fact that land-dwelling vertebrates have an intrinsic pathway does NOT negate the existence of an irreducible core. Incidentally, this diagram contains all of the components which Behe said are part of the irreducible core. So Miller in no way refuted Behe.

As we saw above, the bit about “land-dwelling vertebrates” all having the same system is wrong, since chickens (at least; we have no other complete diapsid (“reptile”/bird) genomes yet) lack Factor XII/Hageman factor.

Other stuff: Luskin’s essay, part 2

Part 2 of Luskin’s essay contains reiterations of a bunch of stuff we’ve dealt with already, e.g.:

Casey Luskin writes:

In Part 1, I showed how Ken Miller misrepresented Michael Behe’s arguments about the irreducibility of the blood-clotting cascade to Judge Jones during the Kitzmiller v. Dover trial, such that Judge Jones wrongly ruled that “scientists in peer-reviewed publications have refuted Professor Behe’s predication about the alleged irreducible complexity of the blood-clotting cascade.” To briefly recap, Miller told Judge Jones that Behe’s discussion of the blood-clotting cascade in Darwin’s Black Box was “essentially identical” to the discussion of the blood-clotting cascade in Of Pandas and People, implying that any critiques of Pandas also applied to Behe. But unlike Pandas, Behe explicitly did not argue that all of the components of the blood-clotting cascade were required for it to function properly.

As we saw before, Behe himself wrote the Pandas discussion as well, and testified on direct that the treatments were “essentially the same.”

Luskin then reiterates the bicycle/unicycle analogy which we already dealt with. Then:

Casey Luskin writes:

Miller assumed that since jawed-fish / dolphins lack certain components found in the intrinsic blood clotting pathway of land-dwelling vertebrates, that therefore the land-dwelling vertebrate blood-clotting cascade is NOT irreducibly complex. But Miller didn’t cite an experiment that demonstrated that the blood-clotting cascade of land-dwelling vertebrates doesn’t require the intrinsic pathway. In other words, he never tried removing a wheel from the bike, which is the proper experiment to show that a bike is reducibly complex. Instead, Miller effectively said unicycles only have one wheel, so bikes don’t need both wheels. It’s an experimentally deficient and logically flawed argument.

As it happens, nature did this experiment for us. Chickens are a “land-dwelling vertebrate” that are missing Factor XII, just like whales/dolphins. It appears that Factor XII is only useful sometimes, in some species. It is probably required/nearly required in some organisms, helpful in others, and useless/harmful in others. A few years back I looked at the literature on Hageman factor. I don’t have all the papers handy, but my recollection is that (1) a deficiency in Factor XII is not a big deal even in humans (wikipedia confirms), except (2) it may be fairly important during child birth, which can be a high-bleeding event (although this study contradicts that idea), and (3) it can trigger unhelpful clotting in some situations. In fact, one of those situations appears to be high pressure, a situation often experienced by diving mammals like whales. This is suggested as the explanation for why Hageman factor was lost in whales in the 1998 Semba et al. paper on the Hageman factor pseudogene:

Saito et al. reported that all marine mammals examined by them did not possess coagulation activities equivalent to Hageman factor and prekallikrein [8]. Their explanation of this observation is as follows: the lack of these initiation factors of the intrinsic blood coagulation pathway must be advantageous to the marine mammals in prevention of disseminated intravascular coagulation syndrome which might occur in the semi-static state of blood circulation in the skin and the lungs under high hydrostatic pressure.


Participation of the intrinsic blood coagulation pathway in the disseminated intravascular coagulation syndrome has been suspected. The whale may dive to depths of 1000 m, where a semi-static state of blood circulation in the skin and the lungs might occur due to the high hydrostatic pressure. The fact that the initiation factor of the intrinsic coagulation system is not produced in the whale would support the clinical suspicion about the participation of this system in the disseminated intravascular coagulation syndrome. (Semba U, Shibuya Y, Okabe H, Yamamoto T., 1998, Whale Hageman factor (factor XII): prevented production due to pseudogene conversion. Thromb Res. 1998 Apr 1;90(1):31-7.)

Obviously this doesn’t explain the absence of Factor XII in chickens (their ancestors apparently lost it at some point for unknown reasons), or in fishes (which never had it). I doubt pressure is an issue with fishes, since they do not breathe air (which is what compresses with depth, not water) and their blood circulation presumably does not become static during dives. It may be that fishes can get away with a simpler system, or, more likely, the various fishes have all sorts of modified systems, some simpler and some more complex than the mammalian system, depending on their size, their activity level, their risk of potentially survivable injury, etc. We won’t know until the blood clotting genes (at least) of dozens or hundreds of species have been characterized.

Casey Luskin writes:

Arguments about the irreducibly complex core aside (see Part 1), the entire land-dwelling vertebrate system might still be irreducibly complex, for the fact that some vertebrates lack factor XII (called Hageman Factor) in their blood-clotting cascade, or even lack the entire intrinsic pathway, in no way implies that humans (and other land-dwelling vertebrates) don’t require all of these components for their blood to clot. Miller didn’t cite the proper experimental data to show that the land-dwelling vertebrate blood-clotting cascade is reducibly complex.

As we saw, one can cite wikipedia for how Hageman factor isn’t even required in humans. We also saw that the advantageous or deleterious nature of a part of a system can change depending on the organism’s environment and habits.

Luskin, desperate to maintain the credibility of Behe and IC, tries to distract us by arguing that the proteins not required in fishes or whales might be required in other critters. And this might be true. But it is beyond obvious, or should be, that the original argument was about whether or not these systems could evolve. Behe said that they couldn’t, because “any precursor to an irreducibly complex system that is missing a part is by definition nonfunctional” (Behe 1996, p. 39). If there are blood-clotting systems that work without all of the allegedly “required parts”, this particular argument that the system couldn’t have evolved, because simpler systems couldn’t work, is sunk.

Casey Luskin writes:

He just cited a unicycle, which as we’ve seen, isn’t sufficient to demonstrate the point he’s trying to make. Miller could only make this argument if he presented actual knockout experiments on the blood-clotting cascade of land-dwelling vertebrates that removed certain components from the blood-clotting cascade, and found that the blood still clotted properly. But Miller didn’t do anything like that.

I’m pretty sure this only makes sense to Luskin because he is an old-earth creationist and doesn’t accept common ancestry beyond the “created kinds” of Genesis, and for him the ICness of one system in organism A might have been miraculously created that way, even if miraculously created organism B doesn’t need all of the parts.

Anyway, we have already dealt with Luskin’s objection to the comparative argument, so I won’t belabor it. However, notice this point: this confirms that Luskin adheres to the “knockout definition” of IC. If you knock something out and the system breaks, then it is part of the IC system. The problem is that there are any number of cases in blood-clotting, the flagellum, etc., where the same protein is required in one the system of one organism, helpful in the system of another organism, and absent in the system of a third organism. Why would this be, if the ID advocates were correct, and irreducibly complexity blocks gradual evolution? This state of affairs is easy to understand if many “IC parts” evolve by first being selected as being useful, and later become necessary. (This doesn’t explain every new part, e.g. cooption is needed; but it is a crucial piece of the overall explanation in every system I’ve looked at).

Casey Luskin writes:

In fact, no one has demonstrated an evolutionary pathway from the blood-clotting cascade of jawed fish to the blood-clotting cascade of land-dwelling vertebrates. In contrast, as Behe observes in Darwin’s Black Box, the blood-clotting cascade is irreducibly complex with respect to the extrinsic pathway and everything after the intrinsic and extrinsic pathways converge, i.e. from factor X onward (see the red box in the diagram below). As usual, Darwinists have used bogus comparisons to other living organisms to find other systems which COULD represent islands along an evolutionary pathway, but they need MUCH more evidence to demonstrate that the pathway exists.

This is the old “you have to give us impossible amounts of evidence or we’re going to maintain that a miracle (err, we mean ID) did it, even though we have no details or tests whatsoever of our explanation” routine, which was dealt with most spectacularly during the immune-system cross in Kitzmiller. For blood-clotting, it is enough to point out, as Ken Miller does, that Doolittle and coworkers have explained many of the blood-clotting factors through several copying and rearrangment events, and these explanations have been tested and confirmed with new genomic results. Must of this literature existed in 2004, before the Kitzmiller case, but check out Ponczek et al (2008):

It has long been apparent that FXI [plasma thromboplastin antecedent or PTA] and PK [prekallikrein] are so similar - both in sequence and domain arrangement - that the duplication that gave rise to these paralogs.2 must have been quite recent, and it was suggested that one or the other might be absent in non-mammalian vertebrates [13].


Genes for FXI and PK in vertebrate genomes

A single paralog of FXI and PK occurs in frog, chicken, and platypus, suggesting its first appearance among early tetrapods (Table 1). In contrast, the opossum genome has genes for both PK and FXI, indicating that the gene duplication leading to separate factors occurred early in mammalian evolution but after the divergence of monotremes (platypus).


  1. Doolittle RF, Feng DF. Reconstructing the evolution of vertebrate blood coagulation from a consideration of the amino acid sequences of clotting proteins. Cold Spring Harb Symp Quant Biol 1987; 52: 869-74.

(Ponczek MB, Gailani D, Doolittle RF. (2008). “Evolution of the contact phase of vertebrate blood coagulation.” J Thromb Haemost, 6(11):1876-83.)

This paper cites dozens of other papers, most of them directly on the evolution of blood-clotting.

Finally, Luskin goes off the deep end:

Casey Luskin writes:

Revealing Differences Between the Blood-Clotting Cascade of Land-Dwelling Vertebrates and Dolphins / Jawed Fish?

During the Kitzmiller trial, Ken Miller discussed the fact that dolphins (like jawed fish) lack factor XII (Hageman factor) in their blood-clotting cascade. This is interesting: Dolphins are supposedly descended from land-dwelling vertebrates (which have factor XII) but their condition is like jawed fish, which lack factor XII. This implies that there may be functional constraints on water-dwelling vertebrates to have a different activation pathway than land-dwelling vertebrates.

We have already discussed how pressure issues are the likely cause of factor XII/Hageman factor loss in ceteceans, and how the existence of chickens without Hageman factor disproves any idea that The Designer was Designing the presence/absence of Hageman factor to correlate with whether or not the critters in question were aquatic.

Conclusion: The Hageman Factor (FXII) Pseudogene and Main Point

Casey Luskin writes:

Darwinists like Ken Miller view the dolphin’s lack of factor XII as a case of convergent evolution, but we might also see it as evidence of a functional constraint or a case of common design. The fact that jawed fish lack factor XII is not necessarily evidence that their blood-clotting cascade was a “primitive evolutionary precursor” to the land-dwelling vertebrate blood-clotting cascade, but evidence of a functional constraint for water-dwelling vertebrates–a constraint which is confirmed in that dolphins also lack factor XII.

This is an interesting issue that will require further research to sort out. In the mean time, any claims that Miller refuted Behe–or even Pandas–appear to be premature.

Um, no. Luskin is apparently unaware of the fact (which Miller mentioned) that whales/dolphins have a pseduogene for Hageman factor. (Or if he is aware of the pseudogene, then he thinks The Designer created whales/dolphins from scratch, complete with a pseudogene for a protein they don’t use, which furthermore is closely related to the corresponding functional artiodactyl Hageman factor gene!) The pseudogene is direct proof that their ancestors had Hageman factor, but lost it. [4] Fish never had it, so this is in no way a case of convergence. What we do have here, however, is an excellent case of someone, namely Casey Luskin, inserting the miraculous intervention of a “common designer” where he has gaps in his knowledge. This is exactly the problem with ID/creationism – invoking God into gaps in knowledge is pretty troublesome, but creationists do something even worse. They insert God into gaps in their own knowledge, assuming, usually without even a vaguely serious attempt at a literature search (!!!), that whatever tidbits of biology they happen to have picked up represent the sum total of scientific knowledge on a topic. This is, I think, why they so often stay ignorant, even when, as Luskin did, they have had the whole thing explained to them before.

There is, though, one thing even worse than inserting God into gaps in one’s own knowledge: inserting God into the gaps in children’s knowledge, via the government schools. This is the end result of the political machinations of creationists when they succeed. They take their own ignorance, and force it into the next generation via the public schools, explicitly or implicitly filling in the holes with God. This should and does offend anyone who respects science, or God, or both.


[1] The authorship of Pandas is complex. In addition to Davis and Kenyon (and Behe) these are the authors we know about: Young-earth creationist Nancy Pearcey was the main author on the long “Overview” chapter, which is not obvious in the book, but confirmed by several obscure sources, especially if you notice that much of that content is replicated in her articles in the Bible-Science Newsletter. Charles Thaxton (listed) and Jon Buell (unlisted) edited the book; Thaxton wrote a philosophical overview for the first edition; this was replaced with a piece by Stephen Meyer (later directory of the ID program at the DI) and Mark Hartwig in the second edition. This piece is online at the DI and various other ID websites.

[2] I have done this for the flagellum. It is entertaining to watch the “number of required proteins” drop from 50, to 40, to 30, to 20 or so between 1996 and the present, as the ID advocates get burned again and again with the higher numbers as new research, or more commonly new reviews, come out.

[3] There are still pretty major issues even once this silliness is dealt with; see Sober’s book Evolution and Evidence. There is all sorts of fuzziness in what is meant by “function” even in standard biology. But for the sake of argument I think it is fair for ID advocates to use the term “function” in the common-sense way that biologists use it; here I am just objecting to circularity that is included when the parts are included as part of the definition. Natural selection doesn’t “care” what the parts are, only the function that results.

[4] Parenthetically, wonder of wonders, the whale/dolphin psuedogene for Hageman factor is most similar to the (non-pseudogene) sequences of other artiodactyls, providing independent evidence for the now-accepted idea, long ridiculed by creationists, that whales and artiodactyls are closely related.