Michael Behe: Kafka at the Dover Trial

15 years ago, the biochemist Micheal Behe was a star witness at the Dover Trial. He explains the trial was Kafkaesque, a surreal experience.

@Nlents you will get a real kick out of our back and forth on human evolution. I hope the biologists on the board can help resolve our disagreement about the meaning of exaptation too.

Panda’s Thumb covered this here:


Looking forward to listening!


This should be interesting.


51 posts were split to a new topic: Daniel Arant: Questions about Evolution and Design

I just learned that Behe affirms common ancestry (at least to an extent). This is totally new to me. Does he more or less believe in directed evolution and that all life shares common ancestry? If so, that’s super cool, and I have way more respect for his honesty. I still disagree with irreducible complexity, but it’s definitely a step in the right direction!


I think you did really well @swamidass. And I have to say I think Behe was extremely confused on the topic of exaptation. Clearly the taking of an existing structure and rearranging it so it serves another function is an example of exaptation. Under that clearly canonical definition of exaptation, Michael Behe is demonstrably incorrect to say there have been no examples of exaptation resulting in novel functions and structures in experimental biology.

Just to pick one recent example I wrote about on this very forum, the rearrangement of parts from different genes resulted in a novel chimeric transporter:

Here’s an example where gene fusion of different genes (the exaptation of parts already existing of protein coding genes serving other functions) results in a novel function:

Of course, the repurposing of the promoter for the rnk gene which occurred spontaneously in the Lenski long-term evolution experiment, through gene duplication, resulted in a novel phenotypic function: The Cit+ phenotype. Hence it is another textbook example of exaptation resulting in a novel function(see figure for how this occurred):

I have a post on Panda’s Thumb explaining how experiments have shown that pieces of already existing protein coding genes can be copied and rearranged into new functional genes, which would also qualify as examples as exaptation:


What of his claim that rearrangements are not exaptation?

I think it needs to be said that irreducible complexity is a real thing. Some functions really are irreducibly complex in the sense that, if you remove a part from the system, it stops having some function. For example, if you remove the filament from the bacterial flagellum then it really does stop functioning as a propulsion system. Problem is that doesn’t make it functionless. If you take away the filament, you still have the flagellar type-III secretion system, which works just fine as protein export machine.


Yeah that doesn’t make any sense, he’s just clearly wrong. If you rearrange parts and you get some new function, then those parts really have been exapted to perform a new function. No ifs or buts that I see, he’s just wrong. I’m 100% with you on this.


Honestly was surprised by that one. He literally says that new function can arise by rearrangement, and some how this is meant to be dismissive of evolutionary processes?


I agree. Thanks for the important clarification! Also, even if modern systems are irreducibly complex, that doesn’t mean the precursorial systems were


Well, not very much of the discussion there was actually about Kitzmiller. But to the extent it was, I mostly find myself strangely in agreement with Behe, except to the extent that he disparages the intellectual capability of Judge Jones.

I think that it is a mistake to assume – as Behe seems to have assumed, in going into the Kitzmiller litigation – that litigation in a common-law court is a process meant to discover some sort of ultimate truth. It is a mistake to assume that even those who practice law believe this – indeed, they believe it less than anyone else does. Litigation is war, and as Sherman said, “War is all hell.”

Understand this: we have, in contrast to much of the civilized world, a deep commitment to the notion of an adversarial system of justice. The civil law model, of an inquisitorial system where judges take an active role in establishing the facts and the law, has certain arguable merits, and you’ll find that system in most European countries. But our system is and has always been one based upon combat: the notion is that a rule-bound, refereed combat in which each side strives to put forth the best case it can will tend to produce legitimate results. If you are an expert and you don’t face harsh cross-examination, frankly, that probably is because your testimony isn’t very important. If your testimony is very important, as Behe’s unquestionably was here, then you should expect the sharpest sort of conflict. The judge is only a referee as to the fairness of that conflict, both procedurally (what kinds of questions can be asked?) and substantively (what is to be made of the conflicts between witnesses on both sides?).

Next: does anyone think that debate in a courtroom and deliberation upon scientific expert testimony is a good way to resolve a scientific dispute? No. Nobody thinks that, least of all those who practice litigation. So why do we do it?

We do it because the adversarial system is based upon cases or controversies. Courts do not reach out into the world to find disagreements and resolve them. They take parties who are at loggerheads, one of whom has refused to back down from some stance and the other of whom has decided that, that being the situation, it must after all be war. And the case is ABOUT that conflict. It is about flesh-and-blood people. In this particular case, it was about religious extremists who chose to take a stand, and citizens who decided they weren’t bloody going to take it any longer. That’s the conflict. It is the role of the lawyers, in that conflict, to analyze that conflict, break it down into issues, frame those issues and endeavor, to the extent those issues are issues of “fact” rather than of “law,” to bring evidence to the court. It is then the role of their adversaries to explain away, contradict, and/or discredit that evidence.

Behe talks about Jones being an English major, lawyer, and “failed politician” as though any of these things have anything to do with it – as if to ask how such a miserable creature could be presumed capable of understanding the illuminating insights of a Behe. The fact is that nobody expects a US District Judge to come to a case pre-loaded with expertise bearing upon the very points at issue. I have litigated, before these judges, all of these matters:

  • The extent and nature of floodwater behavior and management in river plains;
  • The uniqueness and patentability of particular modes of creating the timbre of the individual notes on a church organ;
  • The relationship between prison overcrowding and inmate-on-inmate violence;
  • Police approaches to domestic violence, and the effectiveness of those approaches;
  • The manner of termination of a municipal cemetery management contract;
  • The nature and consequences of a collective bargaining agreement governing the closure of a manufacturing plant, as bearing upon the entitlement of then-disabled workers to pension benefits;
  • The scope of visibility afforded a school bus driver by various configurations of mirrors, and the effectiveness of additional mirrors and other equipment in preventing children from being run over;
  • The impact upon fish populations of construction of small run-of-the-river hydroelectric plants on mountain streams;
  • The feasibility of anadromous fish passage through particular types of drainage ways;
  • and so on, and so on.

We do not expect – and could not reasonably expect – our system to have, as its judges, people who come to the bench richly aware of the full range of technical issues with which they may have to cope. What we do expect is that the parties will choose experts well. What we do expect is that judges will be intelligent and patient. And – though the trendline is rather ugly in recent years – for US District Courts the intellectual quality of our judges has been, on the whole, excellent. Our state courts are quite variable, not just because of the quality of the judges but also because they tend to have excessively large dockets and limited staffing as compared to a US District Judge.

Behe’s argument that Jones did not understand the evidence seems to be based primarily upon Jones pulling passages from the writings of the plaintiffs’ lawyers. But these passages, it must be understood, are from the parties’ proposed findings of fact and conclusions of law. These proposed findings and conclusions are REQUESTED by the court from the parties for just this purpose. While on this occasion he did not suggest that this was plagiarism, he and others have done that elsewhere and it is really misleading. The purpose of these proposed findings is for use in the ruling. That’s why the judge asked for them, and that’s how they were used. The fact that they were used in this case, as they are in almost every case that goes to bench trial, is truly unremarkable and says absolutely nothing about whether the judge subjectively understood or did not understand the testimony. The likelihood of the judge having uncritically accepted them is basically nil, and he and his law clerk(s) will have worked on this ruling for quite a while to get it right.

What does the case resolve? That’s the subject of some misunderstanding all up and down the spectrum of views. But it’s fair to say that when the law is confronted with an issue of fact (is ID a scientific theory?) it seldom if ever purports to resolve that fact for all parties, for all purposes, and for all time. Remember: the whole adversarial system is based upon the case or controversy. The system is built to render decisions not primarily for their far-reaching effects but for their proximate effects upon the behavior of the parties. The public sees it differently because the law people tend to hear about is the far-reaching law: Brown v. Board of Education, or Roe v. Wade, or Korematsu v. United States – cases which, for good or for ill, have massive policy consequences and which raise basic issues about human freedom, autonomy and equality. What you don’t hear about is the daily grinding-through of thousands upon thousands of disputes, each of which turns on its own particular details. We have “landmark” cases because sooner or later, amid those undistinguished cases, a conflict arises which encapsulates something with a depth and breadth that – due to our regard for legal precedent – means that it will mean more to those who are in its wake than it means to the very persons involved. That is the exception, not the rule. Let’s talk about what a ruling like Kitzmiller actually means – something liable to be overstated by people from all sorts of points of view.

A final judgment after trial precludes parties from relitigating the facts, except under narrow special circumstances. It establishes, likewise, the “law of the case” which means that (again, barring certain special circumstances) the parties may not reargue the legal issues. It resolves the claims which were presented, and so here it establishes that the defendants did violate the plaintiffs’ civil rights under the First Amendment, and were liable for that violation under the Civil Rights Act of 1871.

But those are the effects upon the parties: upon the Kitzmillers and other plaintiffs, and upon the Dover Area School Board, and upon the individual defendants on that Board. The effects of the decision more broadly are not so strong. Because this is a decision of a trial court, any other trial court in America (state or federal) stands at the same level and is not “bound” by any principle of stare decisis. Another trial court could hypothetically face the EXACT same facts and rule in the other direction. In any case very similar to Kitzmiller, there is no doubt that the anti-ID party will point to it and insist that the same answers should follow; but he will know, when so doing, that the value of the ruling is advisory only. The pro-ID party is free to argue that Jones got it wrong; he is free to argue that newer and better evidence shows that whether Jones was wrong or not, he would be wrong under the evidence we now have; he is free to argue that the circumstances and details of the new case are sufficiently different as to make the Kitzmiller ruling wholly irrelevant.

The law doesn’t presume to decree, for science, what is or is not science. But the law must presume to decree, for the law, what is or is not science, when that issue is squarely put at the center of a case or controversy as it was in Kitzmiller. A court faced with a case where the parties do not put it in issue, or where the merits may be resolved without reaching the issue, may decline to decide it. But when, as here, the parties on both sides are in full agreement that the scientific character of ID is at issue, the court has no choice but to resolve that question on the basis of the evidence presented by those parties.

Whether one considers this process “Kafkaesque” is, I suppose, a matter of opinion, and Behe is certainly entitled to his. But having been before some adminstrative boards now and then in my career who, as a colleague put it, “really put the ‘quasi’ in ‘quasi-judicial’,” I can’t agree that there is anything Kafkaesque – or even terribly unusual – about how this case proceeded. If this is Kafka, well, call me Gregor and watch what you do with the bug spray.


Behe declares that whenever something requires cooperative change then evolution can’t explain it and nobody can explain it and Darwinism fails. He says this after asserting with great confidence that nobody, including Eugene Koonin (why him of all people?), can explain how something like a disulfide bridge can evolve.

If Behe was right, the adaptive immune system would be impossible, as antibodies could not evolve to adapt to novel pathogens, such as in evolutionary arms-races between host and parasites (such as viruses). This is exactly what occurs with every new seasonal flu. We are hit by a novel mutated strain of influenza, so now your existing antibodies have to change and adapt to catch up, and round and round it goes. It seems to work just fine. And they’re not even cooperative, one side is actively evolving away from being recognized by the immune system.

But anyway, since Behe seems obsessed with the ability of proteins to bind to each other and the supposed impossibility of being able to explain how this can evolve, here’s a recent paper where scientists physically and population-genetically model the evolution of binding spots in protein complexes with multiple subunits:

Zabel WJ, Hagner KP, Livesey BJ, et al. Evolution of protein interfaces in multimers and fibrils. J Chem Phys . 2019;150(22):225102. doi:10.1063/1.5086042


A majority of cellular proteins function as part of multimeric complexes of two or more subunits. Multimer formation requires interactions between protein surfaces that lead to closed structures, such as dimers and tetramers. If proteins interact in an open-ended way, uncontrolled growth of fibrils can occur, which is likely to be detrimental in most cases. We present a statistical physics model that allows aggregation of proteins as either closed dimers or open fibrils of all lengths. We use pairwise amino-acid contact energies to calculate the energies of interacting protein surfaces. The probabilities of all possible aggregate configurations can be calculated for any given sequence of surface amino acids. We link the statistical physics model to a population genetics model that describes the evolution of the surface residues. When proteins evolve neutrally, without selection for or against multimer formation, we find that a majority of proteins remain as monomers at moderate concentrations, but strong dimer-forming or fibril-forming sequences are also possible. If selection is applied in favor of dimers or in favor of fibrils, then it is easy to select either dimer-forming or fibril-forming sequences. It is also possible to select for oriented fibrils with protein subunits all aligned in the same direction. We measure the propensities of amino acids to occur at interfaces relative to noninteracting surfaces and show that the propensities in our model are strongly correlated with those that have been measured in real protein structures. We also show that there are significant differences between amino acid frequencies at isologous and heterologous interfaces in our model, and we observe that similar effects occur in real protein structures.

Doesn’t seem impossible to me. Call me strange.


Exactly. There are certainly irreducibly complex structures - but their existence does not mean that they could not have evolved.


This struck me as indicative of an unreasonable expectation on Behe’s part. Almost all judges are lawyers (particularly federal judges), and most lawyers would come from a liberal arts background rather than a serious scientific background. If you don’t want somebody with such a background passing judgement over your scientific claims then quite simply don’t become an expert witness. I would note that after Dover, Behe participated as an expert witness in a second case, ACSI v. Roman Stearns, with similar results (so for all his complaints about Dover, it did not seem to immediately alter his view of court cases).

If you want your claims to be judged by scientists, the obvious way to do so is to present them for scientific peer review, which Behe seems to have largely avoided. This route would also be useful even when you intend to present your claims in court, as having passed that review gives your testimony more weight, and can be used to help defend your claims when challenged.


Well, yes. And another point: it’s the expert’s job to help the judge understand the testimony. Ordinary fact witnesses are there to tell the court the sorts of things that happened in the course of their own experience and observations. The rationale for having expert testimony is that sometimes it is not enough to HAVE the evidence; sometimes you need to have someone to help EXPLAIN the evidence. If judges were supposed to be the modern version of a Renaissance man, and bring their OWN expertise to every technical issue, there’d be no reason for Behe to be there at all.

Now, it may not be the fault of an expert witness when his testimony fails to make the judge see things the expert’s way. Judges can miss things, for sure, and can get it wrong. But what was Jones to do, on THIS testimony? Go with the guy who affirmed a definition of science so broad that it swept in astrology? I cannot see a way to say that Jones erred in rejecting that. I cannot understand the mindset of anyone who thinks that that particular passage in the testimony was not a five-alarm fire. Yes, long explanations and apologia for what Behe was really getting at, and what he really meant, and why it shouldn’t have mattered, can be formulated and offered, but really: a man who expects to be regarded seriously figures out how to distinguish his views from those of astrologers when a question like that comes up. I think that the reality is that Behe’s testimony was enormously helpful; but it was helpful to the plaintiffs rather than to the Board. Having had the experience of having an expert implode similarly on the stand at trial once, I practically have PTSD every time I read that part of the transcript. To get something worse they’d have had to call a YEC geologist.

I bristle at the idea that when a judge rejects the notion of astrology and its kin as sciences, this can only reflect his intellectual naivety, the sort of thing one can deplore in failed politicians, lawyers, or, worse, English majors. If these notions held sway in the scientific community instead of being the province of a handful of angry cranks, I think I would start getting my science from English majors instead.


Yes, I cannot help but suspect that Behe’s protestations are more about attempting to justify himself in his own eyes (turning himself from a witness who performed poorly on the stand into the victim of a bad judge) than about any genuine flaw in the trial process.

I also cannot help but feel that he shouldn’t have been wandering into the weeds of “what is science?”, given that he isn’t a Philosopher of Science. Maybe Stephen Meyer was originally meant to cover that, but when he became unavailable, it would’ve been better to get another philosopher in than for Behe to try and cover a problematical claim well outside his area of expertise.

I got the impression a few years ago that the Astrology issue may have gotten nailed down by the Plaintiffs’ lawyers at the deposition stage, but I don’t think I’ve ever actually seen a transcript of his deposition for that case, so can’t know if that’s true (or why they weren’t able to walk it back at trial, if that was the case).

I seem to remember the same thing happening in ACSI v. Roman Stearns. As in Dover, the judge cited Behe in deciding for the other side.

Yeh, I was on the opposite side of such an implosion 17 years ago. Our opponent’s expert witness literally wrote the book on the topic, but failed to allow for data idiosyncracies that played havok with the standard technique (which meant you ended up measuring the ‘curvature’ of a stepwise ‘curve’), and then imploded on the stand (I was not there in person, but read the transcripts – and my colleagues who were there said the transcripts understated the carnage).

I saw some years ago Behe trying to claim that he was talking about how Astrology was viewed historically. But (from memory of looking closely at it at the time this claim was made) the question was put to him in the present tense, and there was nothing in his answer that altered that context.


I think that’s right. And what that means is that the plaintiffs’ lawyer, when asking the question, expected to have to deal with the “I’ve since thought more about this” modified response and the squaring of that response with the deposition. I suspect he was a bit surprised to see it repeated without anything to wall it in or explain it. When a man catches fire at the deposition, you do expect that he will have an extinguisher in hand at trial.

I would say that the normal response of plaintiffs’ counsel to that exchange that the deposition would be to say “you’re fired.” However, in this case the expert list had been winnowed down by the DI’s panicked exit, and there was no possibility of trying the case without him.

Yes, I think I recall seeing that, too. It wouldn’t have helped him very much at trial but it would have helped a bit. The difficulty is that he’s then saying very little more than that a hypothesis can be of a scientific character without being true. Sure, but who cares? We don’t teach kids that walruses are insects just because this is a proposition one can formulate and test. I can’t think it would have done him a lot of good.

1 Like

A post was split to a new topic: Questions about Behe/Swamidass at TAMU