Kitzmiller v. Dover Area School District

THE COURT: All right. Mr. Muise, you may continue.

MR. MUISE: Thank you, Your Honor.

BY MR. MUISE:

Q. Dr. Behe, Dr. Miller severely criticized Pandas for its treatment of the topic of protein sequence similarity. Do you agree with his assessment?

A. No, I don't.

Q. And I would ask you to explain why not?

A. On the next slide, we see one of Professor Miller's slides, the first, I think, in his sequence where he very severely criticized the book Of Pandas and People for its treatment of the question of why similar proteins in separate organisms have the differences in their sequence that they do.

And on the next slide, this is again a slide from Professor Miller. He reproduces a figure from Pandas which shows -- it's hard to read on here -- that the difference in the number of amino acids of a protein called cytochrome c, which is a small protein which is involved in energy metabolism and which has about 100 amino acids in it, the difference between that protein which occurs in fish is about 13 percent.

About 13 amino acids differ between the fish cytochrome C and frog cytochrome C; and about 13 or so between bird and fish cytochrome C; and about 13 between mammalian cytochrome C and fish cytochrome C. So that remarkably, the proteins in these different organisms all seem to have roughly the same number of differences, although the differences are not the same differences, but they have the same number of differences from fish cytochrome C.

And Pandas discusses this in their text. And Professor Miller -- Professor Miller takes Pandas to task because he says that, in fact, this is a well-studied and a problem that has been solved by evolutionary theory. For example, he says, in fact, these sequence differences confirm that each of these organisms is equi-distant from a common ancestor, which is the actual prediction of evolutionary theory.

He has a little tree diagram there, too. But one has to realize that, in fact, Professor Miller is mistaken. Evolutionary theory does not predict that. Or one could say, evolutionary theory predicts that in the same sense that evolutionary theory predicted that the vertebrate embryos, as drawn by Haeckle, should be very, very similar to it; or the prediction of evolutionary theory after newer results came out, that vertebrate embryos could vary by quite a bit; or the prediction of evolutionary theory that the type III secretory system would be a good pre-cursor for the flagellum; or the prediction of evolutionary theory that the flagellum -- or that the type III secretory system might be derived easily from a flagellum.

So, in fact, what we have, I will try to make clear, is an instance where experimental science comes up with data, and the data is attempted to be fit into a framework. But this data was not predicted by any evolutionary theory.

Q. How was Pandas' treatment of this compared with what Dr. Miller found?

A. In my view, Pandas' treatment of this topic is actually much more accurate than Professor Miller's discussion of the same topic in his testimony here. Professor Miller, in his discussion, where he says that, evolutionary theory predicts this remarkable amount of difference, is referring to something, although he does not call it such, something called the molecular clock hypothesis.

And notice that, in fact, in Pandas, on the page opposite to the figure that Professor Miller used in his presentation, there is a section entitled A Molecular Clock where they go through and discuss some issues with it, which I will talk about later on.

Q. Just to be clear for the record, the diagram, figure 9 that you've been referring to that Dr. Miller cited in his testimony, appears on page 38 of Pandas, is that correct?

A. Yes.

Q. And the discussion of the molecular clock appearing on the subsequent page appears on page 39 of Pandas, as indicated in this slide, is that correct?

A. That's correct.

Q. Do you have some slides and discussion as to how this molecular clock problem is treated in the science community?

A. Yes, I do, and it will probably take about 10 minutes or so to go through it. So please be patient. But here is a cover of the Biochemistry textbook that I referred to frequently here by Voet and Voet, which is used in many universities and colleges across the country.

And they have a section on the molecular clock hypothesis and on cytochrome C in which they discuss these issues. Let's imagine -- I'm going to try to explain a molecular clock. Let's imagine that these lengths of time -- these lines represent time. And down at the bottom of the screen is a time -- a distant time ago, and up at the top is modern time.

And the branches here represent events in the course of life where a population of organisms split into two -- split into two, and one branch went off to form one group of organisms and another group went off to form a different type of organisms.

Q. If I might just interrupt briefly. You're referring to a phylogenetic tree that has vertical lines that branch off to each other, and that's what you're referring to the vertical lines running, two at the top of the diagram, and then they branch off into different sections?

A. That's correct. That's exactly right.

Q. Could you continue, please?

A. Yes. So, for example, at this branch, a population of organisms split off that went on to become plants, and at this branch, a population split off which went on to become animals.

Now I suppose that before any split in the population, the pre-cursor population organisms had a cytochrome c with a certain sequence. We'll say there was a hundred letters. Just think of a string of a hundred letters; Z, Q, A, L, W.

Now, however, when we get to this branch point, we have a group of organisms going off to form the animals, another going off to form the plants. They no longer interbreed, and so that string of a hundred letters representing cytochrome c can't accumulate mutations in it separately.

So, for example, suppose once every year or so, the cytochrome c in the branch that is forming the plants suffered a mutation, so that one of those letters changed from what it had been. And similarly, in the branch going off to form the animals, once every hundred years or so, one of those letters changed into something.

Not necessarily the same. Maybe a different one. So that after a while, those two sequences would be different. And suppose every hundred years, that happened, one change, one change, one change, and so on. After a while, you'd start to accumulate a number of changes.

Now further suppose that along the line to animals, the population of animals split into two, one line leading to, say, insects, and another line leading to mammals. Now you could have the same thing with the cytochrome c sequence that had been mutating all along, but now they split into two populations, and now these two populations also begin to accumulate mutations independently.

But notice here, they start right at the branch point with the same sequence. But after, say, a hundred years, this will have one difference with what it had at the beginning. This one will have one difference, too. And they don't necessarily have to be the same difference.

So they'll start to accumulate differences with each other between, say, the branch leading to the insects and the branch leading to the mammals. Now here's the point. Any sequence along this branch should have accumulated the same number of sequences between any sequence on this branch.

So that the number of differences between insects and plants should be roughly the same between, as that between mammals and plants. Any animal and any plants should have roughly the same number of differences. Whereas between subgroups of animals that have split off from each other earlier than animals did from plants, they will have had less time to accumulate differences in their amino acid sequences. And so they will have -- so they will have fewer differences.

Q. You mean, if they split off later. You said, earlier. They were split off later, correct?

A. Thank you. Yes, later. So Professor Miller has, I believe, this sort of model in mind, which is commonly -- which is a common way of thinking of these things in science.

So the idea is that, since fish branched off from those other groups of vertebrates, mammals, birds, and so on, the fish, under this model, would be expected to have the same number of differences in their amino acid sequences between themselves and all those other vertebrate groups.

Q. So here you have plants splitting off at the same time as the insects or you have the same -- you have the same connection between insects and plants as plants and mammals?

A. That's right. So the critical point is that, the difference between animals, any animal group like mammals and plants and insects and plants, they should have the same difference between animals and plants, no matter what the subgroup of animals.

But between animals which branch off -- groups of animals which branched off at an earlier -- or from each other earlier to the current time, they would have less time to accumulate differences. And I believe this is what Professor Miller had in mind.

However, this model has some difficulties with it which are well recognized and have been discussed in the literature for over 40 years. For example, I said, suppose every hundred years or so, a mutation occurred. Okay. Well, suppose that in this branch, every hundred years or so, a mutation occurred. But in this branch, suppose a mutation occurred every 50 years.

And suppose when these split, the mutation rate again changed somewhat. Now you would not expect this nice, neat pattern to occur. Now you would expect a jumble. It's not quite clear what one might expect. And it turns out, that's a real problem because it's thought that most mutations accumulate in a lineage when an organism reproduces.

When an organism reproduces, the DNA in it has to be replicated, and that gives a chance for mutations to come into the DNA. But different organisms can reproduce at greatly differing rates. For example, a fruit fly might have a generation time of two weeks, and an elephant might have a generation time of 20 years.

So if the number of mutations that a protein or gene underwent was proportional to the number of generations, you might expect a lineage with quickly reproducing organisms to accumulate mutations much more quickly, and the one with slowly reproducing organisms to accumulate more slowly.

And I believe this is -- on the next slide, there shows discussion from the Biochemistry textbook explaining exactly that point. Let me quote from it. Quote, Amino acid substitutions in a protein mostly result from single base changes in the gene specifying the protein. If such point mutations mainly occur as a consequence of errors in the DNA duplication process, then the rate at which a given protein accumulates mutations would be constant with respect to numbers of cell generations.

Not with time. With numbers of cell generations. If, however, the mutations process results from a random chemical degradation of DNA, then the mutation rate would be constant with absolute time. So here's this complication. If most mutations occur during replication, you wouldn't expect this difference that we see in cytochrome c.

If, for some reason, mutations occurred constant with time, well, then you might expect that. But the problem is, we know of no reason why that necessarily -- that has to be so, why a mutations have to -- would have to occur constant in time.

Q. Is there a problem in addition to this generational rate change?

A. Yes, that's one complication, but there's another one as well. And that's that, this so-called molecular clock seems to tick at different rates in different proteins. And this is an illustration again from the Biochemistry textbook that applies to this point.

On the bottom, the X axis, this is time. This is 200 million, 400 million, a billion years, and so on. This is number of -- or percent amino acid sequence difference. And the idea is that, here's the line for cytochrome c.

Organisms which diverge about 200 million years ago have these many sequence differences; about 400 million years ago, have these many, and so on. Look at how nice and neat that is. However, for another protein, hemoglobin, the molecular clock seems to tick faster. For the same amount of time, hemoglobin has maybe twice as many mutations.

Another region of a protein called a fibrinopeptide seems to accumulate mutations extremely rapidly. And a fourth protein, if you can look at the bottom of the figure, it's hard to see, for something called histone H4, barely accumulates any mutations at all. Organisms in very widely separated categories have virtually identical histone H4's.

Now to resolve this problem, it was postulated that perhaps this has to do with the number of amino acid residues in a protein that are critical for its function. Perhaps in some proteins, you know, most of the amino acid residues cannot be changed or it destroys the function and would destroy the organism.

And in others, maybe some can be changed, but not others. And so you can change those. And perhaps in another group, almost all of them can be changed without really affecting the function. And so that's an interesting idea. But there are also difficulties with that because, under that model, you would predict that if you changed the amino acid sequence of histone H4, then that should cause problems for an organism, because all of its, or most of its, or practically all of its amino acids are critical for function. But experimentally, that is not supported, as shown on the next slide.

Q. Is this -- so you've done work in this area with the histone H4 and the molecular clock?

A. Yes, uh-huh. I've written this commentary in 1990 in a journal called Trends in Biochemical Sciences, commenting on the work of somebody else who experimentally took an organism called yeast into the lab and altered its histone H4 and actually chopped off a couple amino acids at the beginning portion of that protein.

And when he looked, it seems that it didn't make any difference to the organism. The organism grew just as well without those mutations, which is surprising, which is not what you would expect if all of those residues were critical for the function of that protein, histone H4.

Later on, in the year 1996, I and a student of mine, Sema Agarwal, we were interested in this problem of histone H4 and molecular clock, and so we experimentally altered some amino acid residues into protein and changed them into different amino acids, with the expectation that these might destroy the function of the protein. But it turned out not to.

These positions, these amino acids could be substituted just fine, which is unexpected, and which kind of complicates our interpretation of the molecular clock hypothesis. So there are two complications; complications upon complications.

One, we would expect the number of mutations to accumulate with generation time, but it seems to accumulate, for some unknown reason, with absolute time. And the second is that, proteins accumulate mutations at different rates. We would expect that it would have to do with how vulnerable they are to mutations, and mutations might destroy the function of one protein that evolved slowly, but that is not experimentally supported.

Q. Now has this problem been discussed in the scientific literature?

A. Yes, this has been continuously discussed ever since the idea of the molecular clock hypothesis was first proposed in the early 1960's by two men named Emile Zuckerkandl and Linus Pauling. And here are a couple of papers which deal with the difficulties of the molecular clock hypothesis.

Here's a recent one, Gillooly, et al, published in the Proceedings in the National Academy of Sciences, entitled The Rate of DNA Evolution, Effects of Body Size and Temperature on the Molecular Clock. In this publication, they say that, in fact, the size of an organism and temperature can affect how fast or how slow this clock might tick.

Francisco Ayala has written on this frequently. Here's one from 1997. And I should say, Francisco Ayala is a very prominent evolutionary biologist. He wrote an article in 1997 entitled Vagaries of the Molecular Clock. And I think the title gets across the idea that there are questions with this hypothesis.

And in 1993, a researcher named Tomoka Ohta published an article in the Proceedings of the National Academy of Sciences entitled An Examination of the Generation-time Effect on Molecular Evolution in which she considers exactly that complication that the textbook Voet and Voet pointed out, this generation-time effect.

You know, why shouldn't organisms that reproduce more quickly accumulate more mutations. I have another slide just from one more recent paper. This paper by Drummond, et al, is entitled Why Highly Expressed Proteins Evolve Slowly. And it's referring to the sequence evolution that I've been discussing.

It was published in the Proceedings of the National Academy of Sciences, and this was from an online version. This is so recent that I don't think it has yet appeared in print. The point I want to make with this is that, these people treat this question as a currently live question.

They start off by saying, a central problem in molecular evolution is why proteins evolve at different rates. So that question I was trying to illustrate with histone H4, why does one protein tick faster and another one tick more slowly, that's still -- that is still unknown.

And I think I will skip the rest of this slide and go to the next slide and just point out a couple words here. Drummond, et al, say, Surprisingly, the best indicator of a protein's relative evolutionary rate is the expression level of the encoding gene.

The only point I want to make with this is that, they are reporting what is a surprise, what was not expected, which was not known, you know, 40 years ago, which has only been seen relatively recently. And they say, quote, We introduce a previously unexplored hypothesis, close quote.

And the point I want to emphasize is that, here in this paper published, you know, weeks ago, that they are exploring new hypotheses to try to understand why proteins have the sequences that they do.

Q. So in summary, this protein sequence, the fact that the equi-distant from a common ancestor is not what evolutionary theory would actually predict?

A. That's right. Evolutionary theory makes no firm prediction about this anymore than it makes a firm prediction about the structure of vertebrate embryos.

Q. It's a common understood problem that biologists are trying to resolve at this point?

A. Yes, within the community of scientists who work on this. People have been working on it for decades.

Q. Is this a problem that an American Biology teacher should be aware of?

A. Yes, an American Biology teacher should be aware of it, because an article on this very topic was published in the magazine, American Biology Teacher, a couple years ago, which is put out by the National Association of Biology Teachers.

And the article is entitled Current Status of the Molecular Clock Hypothesis. And one of the first -- this is a red arrow that I added to the figure. One of the first subsections of the article is entitled How Valid is the Molecular Clock Hypothesis? And if you'll advance to the next slide, let me just read the last line from the paper.

The author says, The validity of a molecular clock, except in closely related species, still remains controversial. So the point is that, extrapolating across wide biological distances, such as from fish to other vertebrates, that is controversial.

Maybe similar species, species of mice or some such thing, okay. But when you try to extrapolate further, the model is quite controversial.

Q. How does Pandas then address this issue?

A. Well, I have here the section from Pandas entitled The Molecular Clock where they discuss exactly all these things. They discuss the molecular clock, the standard molecular clock model, the naive molecular clock model, and then they discuss complications with it.

Let me just read this section from Pandas on the molecular clock. They write, quote, Some scientists have suggested that the idea of a molecular clock solves the mystery. The explanation they advance is that there is a uniform rate of mutation over time, so quite naturally, species that branched off from a common ancestor at the same time in the past will now have the same degree of divergence in their molecular sequences.

There are some serious shortcomings, however, with this explanation. First, mutation rates are thought to relate to generation times, with the mutation rates for various molecules being the same for each generation.

The problem comes when one compares two species of the same taxon, say two mammals, with very different generation times. Mice, for instance, go through four to five reproductive cycles a year. The number of mutations, therefore, would be dramatically higher than, say, those of an elephant.

Thus, they should not reflect similar percent sequence divergences for comparable proteins. Besides that, the rates of mutations are different for different proteins even of the same species. That means that, for the molecular clock idea to be correct, there must be not one molecular clock, but thousands.

So let me point out here that, in this section, Pandas describes the simple molecular clock idea that was proposed 40 years ago by Zuckerkandl and Pauling, and then talks about the two complications for the model, which are common knowledge and are taught in basic science texts that deal with this issue, the generation time problem and the fact that different proteins accumulate mutations at different rates.

And as I have shown from the literature I just cited, that continue to be live issues in the scientific community.

Q. In that section you read from on the molecular clock from Pandas are found on page 39, is that correct?

A. Yes, that's correct.

Q. Again, returning to that slide that Dr. Miller presented in his testimony?

A. Yes. I just wanted to go back to that slide where Dr. Miller says -- again, I should say that, in his testimony, which I attended, he, you know, excoriated Pandas on this point. And he says -- on his slide, he says, in fact, the information we have confirms that each of these organisms is equidistant from a common ancestor, which is the actual prediction of evolutionary theory.

And that's simply is incorrect. And in my view, Pandas is treating problems that Professor Miller, treating real live problems that Professor Miller shows no signs of being aware of. So I think a student reading this section would actually get a better appreciation for this subject than otherwise.

Q. Dr. Behe, in Dr. Miller's testimony, he also criticized another example found in Pandas that had a message such as, quote, John loves Mary, written on the beach, would be a sure sign of intelligence.

He claimed that any philosopher, any logician would spot the mistake in logic, because we know a human made that message, and probably made it with a stick, because we have seen such things happen in our own experience. Do you agree with this reasoning?

A. No, I disagree with Professor Miller's reasoning.

Q. And if I can just say, the example that John loves Mary, and we have a slide up, that's on page 7 of Pandas, correct?

A. Yes, that's right.

Q. Again, could you explain why you disagree with this reasoning?

A. Yes. The inference from the -- the inference from the existence of designed objects in the -- in our world of experience to the conclusion of design in life is an example of an inductive inference. And I think I explained earlier that, in an inductive inference, one always infers from examples of what we know to examples of what we don't know.

And the strength of the inference depends on similarities between the, between the inference in relevant properties. For example, in the Big Bang hypothesis, scientists extrapolated, or used inductive reasoning of their knowledge of explosions from our everyday world from things like fireworks and canon balls and so on.

They extrapolated from their experience that the motion of objects away from each other bespeaks an explosion. They extrapolated from our common everyday experience to something that nobody had ever seen before, an entirely new idea, that the universe itself began in something like a giant explosion.

Nonetheless, they were confident that this was a good idea because they thought the relevant property, the parts moving rapidly away from each other, was what we understand from an explosion. And that's how science often reasons.

In the same way, the purposeful arrangement of parts in our everyday experience bespeaks design. Pandas is exactly right, that if we saw such a message on the beach, we could conclude that it had been designed. And William Paley is exactly right, that if we stumbled across a watch in a field, that we would conclude that it was designed, because in each case there is this strong appearance of design from the purposeful arrangement of parts.

Now we have found purposeful arrangement of parts in an area where we didn't expect to, in the very cellular and molecular foundation of life, in the cell. The cell again was not understood in Darwin's day. And it is much better understood now. And from the new information we have, again, we see this purposeful arrangement of parts, and it's -- by inductive reasoning, we can apply our knowledge of what we see in our everyday world to a different, completely different realm.

And so that sort of inference has been done in science throughout the history of science, and it's a completely valid inference for Pandas to make.

Q. Now we've heard some testimony throughout the course of this trial of a program called SETI, S-E-T-I, a project, I believe, that stands for the search for extraterrestrial intelligence?

A. Yes.

Q. Are you familiar with that project?

A. Yes, I am.

Q. Whose project is that?

A. The search for extraterrestrial intelligence is a project that was, for a while, was sponsored by the federal government. It involved scientists scanning the skies with detectors to see if they could detect some electromagnetic signal that might point to intelligence.

Q. Is there a comparison with that project to the discussion you had in here with the John loves Mary on the beach?

A. Yes. Again, if they detected something that seemed to have a purposeful arrangement of parts, if they saw something that bespoke a message, then even though we have had no experience with other entities from off the Earth trying to send us a message, nonetheless, we could still be confident that an intelligent agent had designed such a message.

And again, whenever we see John -- things like John loves Mary, we can be confident of that. And when we see the purposeful arrangement of parts in the cell, the argument is that, we can be confident of that, that that bespeaks design as well.

Q. I want to bring this discussion somewhat down to the molecular level, and ask you whether or not new genetic information can be generated by Darwinian processes. And I want to be more specific and ask whether new genetic information can be generated by known processes such as gene duplication and exon shuffling?

A. Well, that's a topic about which you have to be very careful and make distinctions.

Q. Okay. Let's start with the gene duplication. If you could explain what that is in the context of generating new genetic information?

A. Well, gene duplication is a process whereby a segment of DNA gets copied twice or gets duplicated and replicated so that where one gene was present before, a second copy of the exact same gene is now present in the genome of an organism. Or sometimes larger segments can be duplicated, so you can have multiple copies of multiple genes.

Q. Are you saying, duplication, like photocopying, is just making another copy of the gene that was originally existing?

A. Yeah, that's a good point. It's important to be aware that gene duplication means that you simply have a copy of the old gene. You have not done anything new. You've just taken the same gene and copied it twice. So it would be like, like photocopying a page. And now you have two pages, but it's just a copy of the first one, it's not something fundamentally new.

It would be like saying, the example of Pandas here with John loves Mary. If you walked down the sand another five yards or something, and you came across another message that says, John loves Mary, well, that's interesting, but you don't have anything fundamentally new.

Q. Can there be variations though in the duplication of those genes?

A. Well, once a gene has duplicated, then the idea goes that, perhaps one of those two copies can continue to perform the function that the single copy gene performed before the duplication, and the other one is sort of a spare copy.

Now it's available to perhaps undergo mutation, and mutation accumulate changes, and perhaps Darwinian theory postulates. Perhaps it can go on to develop brand new properties.

Q. Does this generate new information? And if you use that John loves Mary example to help explain perhaps?

A. Well, again, you have to be careful. Nobody disputes that random mutation and natural selection can do some things, can make some small changes in pre-existing systems. The dispute is over whether that explains large complex functional systems.

And to leave the world of proteins for a second, to look at John loves Mary, suppose we're looking at the spare copy, and the first copy was continuing to fulfill the function of conveying that information. Well, you know, suppose you changed a letter. Suppose you changed the final n in the word John to some other, some other letter, like r. That would not spell a name in the English language.

So that's kind of an analogy to saying that, you might lose the function of the message in the terms. In the terms of protein, the protein might no longer be functional. But you might get to closeby. You might get to closeby messages. For example, if you deleted the r and the y from the end of Mary, you might get to John loves Ma, or some such thing. But you're not going to get anything radically different from that.

Q. So you are operating with the copy. The copy is operating with those same letters, the John loves Mary, or some variation or deletions of that subset?

A. That's right. A copy is a copy. It's essentially the same thing. And now the big problem that Darwinian processes face is, now what do you do? How do you generate a new complex function?

Q. And that's with gene duplication that we just talked about. Could you explain a little bit about exon shuffling in the context of generating new complex information?

A. Yes, exon shuffling is a little bit more involved. It turns out that the gene for a protein can contain regions of DNA that actually code for regions of a protein interrupted by regions of DNA that don't code for regions of a protein. And the regions that code for the part of the protein are called exons.

Now it turns out that, in cellular processes, similar to gene duplication and other processes, too, one can duplicate separate exons and sometimes transfer them to different places in the genome and other such processes. But to make it more understandable, we can go back to the analogy of John loves Mary.

And in this sense, exon shuffling might be expected to generate something like, instead of John loves Mary, perhaps Mary loves John, or John Mary loves, or something like that. But again, it's kind of a mixture of pre-existing properties, and we're not generatesing something fundamentally new.

Q. So, for example, you couldn't generate Brad loves Jen from exon shuffling using your beach example?

A. No, I hope not.

Q. Do these concepts, particularly gene duplication, exon shuffling, do they have any impact on the concept of irreducible complexity that you've been discussing quite a bit throughout your testimony?

A. Yes. In fact, there is an important point to recognize here. Russell Doolittle knew all about the processes of gene duplication and exon shuffling. And as a matter of fact, in the blood clotting cascade, many proteins look similar to each other, and they're often times pointed to as examples of exon shuffling.

But nonetheless, that knowledge did not allow him to explain how the blood clotting system might have arisen. Again, these are sequence comparisons. And such information simply does not speak to the question of random mutation and natural selection being able to build complex new biochemical structures.

In the same way, the people who are investigating the type III secretory system and the bacterial flagellum know all about gene duplication and exon shuffling. And nonetheless, that information has not allowed them to explain the origin of either of those structures.

So those are interesting processes. And people who are convinced of Darwinian theory include those processes in their theory, but they do not explain -- they do not explain where new complex systems come from. And it's an example of somebody accommodating this information to an existing theory rather than getting information that actually experimentally supports the theory.

Q. So can random mutation and natural selection generate new information?

A. Well, again, that's -- you have to be careful. You can make small changes in pre-existing systems. And that's clearly the case. One can clearly do that. But there has been no demonstration to show that such processes can give rise to new complex systems such as we've been suggesting. And there are many reasons to think that it would be extremely difficult to do so.

Q. Have you prepared some slides with a couple -- several quotes that make this point?

A. Yes, I do. This first one is an excerpt from a paper from John Maynard Smith, which I spoke about earlier, from 1970 entitled Natural Selection and the Concept of a Protein Space. Let me read the first excerpt.

Quote, It follows that if evolution by natural selection is to occur, functional proteins must form a continuous network which can be traversed by unit mutational steps without passing through nonfunctional intermediates, close quote. Again, let me explain.

If you can remember the figure of two proteins binding to each other that I showed in -- I showed yesterday, he is speaking of unit mutational steps in terms of one of those interactions, maybe a plus charge and a minus charge or a hydrophobic group and another hydrophobic group.

And so to get two proteins to -- or proteins to start change into something new and different with different properties, each one of those changes would have to be a beneficial one, or at least not cause any difficulties for the problem. And actually, seeing how that could happen is extremely difficult.

And continuing on this slide. I'm sorry. Could you back up one slide? Thank you. The bottom part of the quotation, he says, quote, An increase in the number of different genes in a single organism presumably occurs by the duplication of an already existing gene followed by divergency. So here, he's kind of describing the standard scenario which -- scenario, which is standard in Darwinian thinking, that one has gene duplication and then divergence of the sequence of a gene, and that gives a brand new interesting and complex protein.

But notice that I, of course, underlined and bolded the word presumably. Well, presumably, you know, is a presumption. And it may be true, and it may not. But presumptions are not evidence. And so in order to support this idea, one needs more than the presumption that it occurs.

Q. Do you have another citation to a science text?

A. Yes, I do. Here's an excerpt from an article by a man named Alan Orr, who is an evolutionary biologist at the University of Rochester. And again, this speaks to the same consideration, that you have to be able to have a pathway that step by tiny step could lead from one functional protein to another.

He says, quote, Given realistically low mutation rates, double mutants will be so rare that adaptation is essentially constrained to surveying, and substituting, one mutational step neighbors. Thus, if a double mutant sequence is favorable, but all single amino acid mutants are deleterious, adaptation will generally not proceed.

Again, this makes the point that, if you only need to change one little step, Darwinian evolution works fine. But if you need to change two things before you get to an improved function, the probability of Darwinian processes drops off dramatically.

If you need three things, it drops off, you know, even more dramatically. And nonetheless, as I showed in that figure of interacting proteins, even to get two proteins to stick together, multiple groups are involved.

Q. Did you write about something similar in a paper?

A. Yes. The paper that I published with David Snoke last year speaks exactly to this topic. It's entitled Simulating Evidence by Gene Duplication of Protein Features that Require Multiple Amino Acid Residues.

And in this theoretical study, we showed that, again, if you need one change, that's certainly doable. If you need two amino acid changes before you get a selectable function, the likelihood of that drops considerably. Three or more, now you're really in the very, very improbably range. So again, gene duplication is not the answer that it's often touted to be.

Q. Can you make an analogy here at all to -- you talked about Maxwell and the ether theory?

A. Yes. When Darwinian -- adherence to Darwinian theory, when they view that there are similar genes in different -- in the same organism, and they infer a process of gene duplication, it is simply their theoretical framework, which is saying, such a process must be important in generating new and complex structures.

That has not been demonstrated. Just like James Clerk Maxwell knew that light was a wave and inferred from his theory that there must be an ether, modern Darwinists infer from something we know, the existence of gene copies to an unproved role of such a process in generating complex biochemical systems.

Q. Now Dr. Miller says that Pandas necessarily rejects common descent, and points to a figure -- I believe it was 4.4 on page 99 -- showing separate lines representing categories of animals rather than a branching tree. Do you regard that as ruling out common descent?

A. No, I don't. And here's a figure that I made up in the upper right-hand corner. It's figure 4.4 from Pandas, which is the figure that Professor Miller showed, which shows straight lines instead of a branching tree, which is the traditional representation of how -- of the fossil record.

Nonetheless, here I regard this as simply trying to describe the data without a theoretical framework, without the branched lines in between. One has to realize that these lines do not occur in the fossil record. These are theoretical constructs.

And how one groups things together is theory building rather than data itself. I viewed this as Pandas trying to describe the data without the framework of the existing theory. And I might add that, this was figure 4.4. And earlier, a couple pages earlier, Pandas describes the traditional interpretation of the fossil record in terms of a branching tree.

And in this section, section 96 through 100, the meaning of gaps in the fossil record, Pandas describes the traditional tree diagram for the fossil record, and then points to statements by biologists, saying that there seem to be difficulties in this sort of representation, and then goes on to discuss what interpretations, what ideas have been offered to try to account for the form of the fossil record.

Pandas writes, Several interpretations have been offered to resolve this problem. That is, that the tree of life doesn't seem to be as continuous as one might expect. Number 1, they say, imperfect record. That is, maybe not all organisms left representative of fossilized specimens. Number 2, incomplete search. And that is, maybe we simply haven't looked in the right places or looked in all the places on the Earth, and maybe when we do, then we will find what we expect to be there.

Number 3, what they call jerky process, or which has been called punctuated equilibrium, which was an idea advanced by Steven J. Gould and Niles Eldredge in the 1970's, whereby it said that the mode or the tempo of evolution is one in which a species or a branch of life stays pretty much constant for a long period of time, and then within a relatively short period of time, large changes occur.

And then fourth, they say, well, perhaps -- they suggest something called the sudden appearance or face value interpretation, saying that, well, maybe if we see the sudden appearance of some feature or organism in the fossil record, then that, in fact, might be what happened.

Nonetheless, as I say, they discuss all of these possibilities, including the standard interpretation. And at the end of the section, they write that, scientists should not accept the face value interpretation of the fossil record without also exploring the other possibilities, and even then, only if the evidence continues to support it.

So as I read this, Pandas is telling students that they should follow the data where the data lead. And if the data lead from this model to another model, or from that model to a second model, then a scientific attitude toward the problem is to follow the data, where the data go.

Q. Dr. Behe, does intelligent design necessarily rule out common descent?

A. No, it certainly does not.

Q. Now we've heard testimony from several witnesses claiming that the theory of evolution is no different than, say, the germ theory of disease, so there's no reason to pay any special attention to it. Do you agree with that?

A. No, I disagree.

Q. And why?

A. Well, in a number of ways, evolutionary theory is unique. It's been my experience that students have a number of misconceptions about the theory. They confuse facts with theoretical interpretations. They do not make distinctions between the components of evolutionary theory.

And perhaps, most strikingly, a number of people have made very strong extra-scientific claims for the implications of evolutionary theory.

Q. Now I just want to return to something you had said about your experience with students. You testified that you teach a course called popular arguments on evolution, is that correct?

A. Yes, that's right.

Q. And you've been teaching that for 12 years?

A. Roughly, yes.

Q. Now are there some standard misconceptions that you can point to about the theory of evolution that you find your students bringing to the class?

A. Yes. In my experience, a number of students come in thinking that, in fact, evolution is completely true; that is, they don't make a distinction between fact and theory, they don't think it will be falsified, or they don't think there's a possibility of it being falsified.

They also confuse various components of evolutionary theory. For example, you can ask a student, you know, why they think Darwinian evolution is correct? And they'll say, you know, because, you know, because of the dinosaurs. And they're mistaking change over time with the question of natural selection. And they will assume that the existence of animals in the past necessarily means that animals in the present were derived from them by random mutation and natural selection.

Oftentimes also, students think that utterly unsolved problems, such as the origin of life, have, in fact, been solved by science. I had students tell me that, gee, it's true, right, that science has shown genes being produced in origin of life experiments. So in my experience, students bring a number of misconceptions to this issue.

Q. One of the first ones you indicated is that they believe that Darwin's theory of evolution is a fact as opposed to a scientific theory?

A. That's right.

Q. Does intelligent design seek to address some of these misconceptions?

A. Yes. Yes, it does. One way is -- one way to address the problem of students not understanding that the distinction between fact and theory is to at least have at least one more theoretical framework in which to treat facts.

If a student has only one theory and a group of facts to think of, it's extremely difficult to distinguish what is theory and what is fact. The little lines connecting various points on, say, a protein sequence comparison are theory, but students can often confuse them, confuse them to be facts.

Q. Do you believe these students will be better prepared if they had learned that Darwin's theory of evolution was not a fact and that gaps and problems existed within this theory?

A. Yes, I certainly do. They would see that, in fact, if you can look at the data in a couple ways, then they'll more easily distinguish data from interpretation or from theory. And if they are aware that there are problems in a theory, then perhaps they won't expect -- they won't, again, confuse it with a fact, they'll understand that there are some problems that are unresolved.

Q. Now you made some indication previously in your answer to my question that there are claims made about the theory that go beyond biology, is that true?

A. Yes, that's certainly true.

Q. And do you have some slides to demonstrate some of those examples?

A. Yes, I have a couple of slides, four slides over -- that point to this. For example, in the high school textbook Biology, which was written by Professor Kenneth Miller and his co-author, Joseph Levine, this is the 1995 version, I think, the third edition, in a section entitled The Significance of Evolutionary Theory, the authors write, quote, The influence of evolutionary thought extends far beyond biology. Philosopher J. Collins has written that, quote, there are no living sciences, human attitudes, or institutional powers that remain unaffected by the ideas released by Darwin's work, close quote.

In another example of the implications, the profound implications beyond biology that some people see for Darwin's theory, there's a section in his book, Finding Darwin's God, A Scientist's Search for Common Ground Between God and Evolution, where Dr. Miller writes that, quote, God made the world today contingent upon the events of the past. He made our choices matter, our actions genuine, our lives important. In the final analysis, He used evolution as the tool to set us free.

So here is a scientific theory which is being used to support the idea that we are free, we are free, in apparently some metaphysical sense, because of the work of Darwin. In another example -- it's just that -- for example, the expert, Professor John Hauck, the theologian from Georgetown University, has written a number of books, including God After Darwin, a Theology of Evolution.

Further example, in -- the evolutionary biologist, Richard Dawkins, in his book, The Blind Watchmaker, writes, Darwin made it possible to be an intellectually-fulfilled atheist.

If I could have the next slide. Thank you. The Darwinian philosopher, Daniel Dennett, who's at Tufts University, has described Darwinism as a universal acid that destroys our most cherished beliefs. And he says, quote, Darwin's idea had been born as an answer to questions in biology, but it threatened to leak out, offering answers, welcome or not, to questions in cosmology, going in one direction, and psychology, going in the other direction.

If the cause of design in biology could be a mindless, algorithmic process of evolution, why couldn't that whole process itself be the whole product of evolution, and so forth, all the way down? And if mindless evolution could account for the breathtakingly clever artifacts of the biosphere, how could the products of our own real, quote, unquote, minds be exempt from an evolutionary explanation? Darwin's idea thus also threatened to spread all the way up, dissolving the illusion of our own authorship, our own divine spark of creativity and understanding.

So again, Professor Dennett sees implications for Darwin's theory that are profound and that extend well beyond biology. Another philosopher by the name of Alex Rosenberg, who's at Duke University, published an article a few years ago in the journal Biology and Philosophy that, quote, No one has expressed the destructive power of Darwinian theory more effectively than Daniel Dennett. Others have recognized that the theory of evolution offers us a universal acid, but Dennett, bless his heart, coined the term.

In short, it, that is Darwin's idea, has made Darwinians into metaphysical Nihilists denying that there is any meaning or purpose to the universe, close quote. So again, a number of philosophers, a number of scientists, and so on, see very, very profound implications in Darwin's theory.

Two more quotations on this last slide on this topic. Larry Arnhart is a professor of political science at Northern Illinois University. He wrote a book entitled Darwinian Natural Right, The Biological Ethics of Human Nature. And in it, he writes -- and in it, he writes the following, that, quote, Darwinian biology sustains conservative social thought by showing how the human capacity for spontaneous order arises from social instincts and a moral sense shaped by natural selection in human evolutionary history.

So let me emphasize that he sees implications for politics from Darwin's theory. And the same -- and a Princeton University philosopher by the name of Peter Singer has written a book entitled A Darwinian Left, Politics, Evolution, and Cooperation. And in it, he writes that we should try to incorporate a Darwinian ethic of cooperation into our political thought.

So the gist of Professor Singer's book is that, Darwinian ideas support a liberal political outlook. And he argues for that. So, again, these -- all of these people see profound implications for Darwin's theory well far beyond biology.

Q. These are non-scientific claims, correct?

A. Yes, that's correct.

Q. Have you come across any similar claims made about, say, the germ theory of disease?

A. I have never seen the germ theory of disease argued to say how we should conduct our political life.

Q. How about atomic theory?

A. I have never seen atomic theory used in such profound senses either. So my point then is that, it is perfectly rationale to treat a scientific theory, which so many people have claimed such profound implications for, to treat it differently from other scientific theories for which such far-reaching implications have not been claimed.

It might be very important, and I think a school district would be very justified to say that, since this particular theory seems to reach far beyond its providence, then we should take particular care in explaining to our students exactly what the data is for this theory, exactly what is the difference between theory and fact, exactly what is the difference between theory and interpretation. And so I think such an action would be justified.

Q. Sir, I want to ask you some questions about creationism as it relates to intelligent design. First of all, let me ask you, does creationism have a popular meaning or is there a popular understanding of that term?

A. Well, again, you have to be careful, because many words in these discussions can have multiple meanings. And if you're not very careful about your definitions, you'll easily become confused.

Creationism -- creationist has sometimes been used, as John Maddox, the editor of Nature, used it, simply to mean somebody who thinks that nature was begun by a supernatural act, by God, and the laws of nature perhaps were made of God, and unfolded from there nonetheless.

Q. That would be similar to Dr. Miller's view towards evolution that he had written in his book Finding Darwin's God?

A. Yes, that seems to be consistent with what he wrote. But nonetheless, in the popular useage, creationism means -- creationist means somebody who adheres to the literal interpretation of the first several books -- or first several chapters of the Book of Genesis in the Bible, somebody who thinks that the Earth is relatively young, on the order of, say, 10,000 years, that the major groups of plants and animals and organisms were created ex-nihilo in a supernatural acts by a supernatural being, God, that there was a large worldwide flood which is responsible for major features of geology, and so on.

Q. Now we've heard different terms; young-earth creationism, old-earth creationism, and special creationism. And you have familiarity with those terms, is that correct?

A. Yes, that's right.

Q. Is intelligent design creationism, whether you call it young-earth creationism, old-earth creationism, or special creationism?

A. No, it is not.

Q. And why not?

A. Creation -- creationism is a theological concept, but intelligent design is a scientific theory which relies exclusively on the observable, physical, empirical evidence of nature plus logical inferences. It is a scientific idea.

Q. Is it special creationism?

A. No, it is not special creationism.

Q. Again, why not?

A. Again, for the same reason. Creation is a theological religious concept. And intelligent design is a scientific idea, which is based exclusively on the physical, observable evidence plus logical processes.

Q. Dr. Miller has made a claim that if the bacterial flagellum, for example, was designed, then it had to be created, and is, therefore, special creationism. Is that accurate?

A. No, that is inaccurate. The reason it's -- again, creation is a theological concept. It is a religious concept. But intelligent design is a completely scientific concept which supports itself by pointing to observable, physical, empirical facts about the world, about life, and makes logical inferences from them.

Q. Does intelligent design require that the bacterial flagellum, for example, instantaneously appear from nothing?

A. No, it does not.

Q. Why not?

A. Because intelligent design focuses exclusively on the deduction of design from the purposeful arrangement of parts. And it says nothing directly about how the design was effected, whether it was done quickly, or slowly, or whatever. So it has nothing to say about that.

Q. Could the bacterial flagellum have been designed over time?

A. Yes, it could.

Q. Does intelligent design require ex-nihilo creation?

A. No, it does not.

Q. Why not?

A. Because again, the term ex-nihilo creation is a theological concept, a religious concept. And intelligent design is a scientific idea that relies on observable facts about nature plus logical inferences.

Q. Is there, again, an analogy you can make here to the Big Bang theory?

A. Yes. Yes, there is. Again, many people, including many scientists, saw in the Big Bang theory something that had theological implications, maybe this, this Big Bang was ex-nihilo creation by a supernatural being. And many people who saw that didn't like that. Nonetheless, the Big Bang theory itself is an utterly scientific theory because it relies on observations, physical observations, empirical observations about nature, and reasons from those observations using logical processes.

Q. Is intelligent design a religious belief?

A. No, it isn't.

Q. Why not?

A. Intelligent design requires no tenet of any particular religion, no tenet of any general religion. It does not rely on religious texts. It does not rely on messages from religious leaders or any such thing. The exclusive concern of intelligent design is to examine the empirical and observable data of nature and reason from that using logical processes.

Q. Now some claim that intelligent design advances a religious belief, that it is inherently religious and not science. Do you agree?

A. No. Again, no more than the Big Bang theory is inherently religious. Although the Big Bang theory and intelligent design might be taken by some people to have theological or philosophical implications, both of them rely on observed evidence, empirical evidence, and logical reasoning.

Neither the Big Bang nor intelligent design relies on any religious tenet, points to any religious books, or any such thing.

Q. Do creationists in the sense that Plaintiffs and, I believe, their experts use in this case require physical evidence to draw their conclusions?

A. No. Actually, it's interesting that one could be a creationist without any physical evidence. One could rely -- a creationist could rely for his belief in creation on, say, some religious text or in some private religious revelation or some other non-scientific source.

So a creationist does not need any physical evidence of the kind that, for example, Richard Dawkins sees in life that leads him to think that life has the strong appearance of design or the kind that David DeRosier sees in the bacterial flagellum. A creationist can believe in creation without any such physical evidence.

Q. Is that different than from a proponent of intelligent design?

A. Yes, that's vastly 180 degrees different from intelligent design. Intelligent design focuses exclusively on the physical evidence. It relies totally on empirical observations about nature. It does not rely on any religious text. It does not rely on any other such religious information. It relies exclusively on physical evidence about nature and logical inferences.

Q. Are intelligent design's conclusions or explanations based on any religious, theological, or philosophical commitment?

A. No, they are not.

Q. Again, can you draw any comparisons between intelligent design and the Big Bang theory in this regard?

A. Yes. Again, the -- both the Big Bang theory and intelligent design may have philosophical or theological implications in the view of some people, but again, both are scientific theories. Both rely on observations about nature. Both make reasoned conclusions from those observations about nature.

Q. Does intelligent design require adherence to the literal reading of the Book of Genesis?

A. No, it does not.

Q. Does intelligent design require adherence to the belief that the Earth is no more than 6 to 10,000 years old?

A. No, it doesn't.

Q. Does intelligent design require adherence to the flood geology point of view which is advanced by creationists?

A. No, it doesn't.

Q. Does intelligent design require the action of a supernatural creator acting outside of the laws of nature?

A. No, it doesn't.

Q. Could you explain?

A. Yes. Making an analogy again to the Big Bang theory, the Big Bang theory is a theory which is advanced simply to explain the observations that we have of nature, and it does so by making observations and making inferences. It does not posit any supernatural act to explain the Big Bang. It leaves that event unexplained.

Perhaps in the future, science will find an explanation for that event. Perhaps it won't. But nonetheless, the Big Bang is a completely scientific theory. Again, intelligent design is a scientific theory that starts from the data -- the physical, observable data of nature, and makes reasoned conclusions from that and concludes intelligent design.

Scientific information does not say what is the cause of design. It may never say what is the cause of design. But nonetheless, it remains the best scientific explanation for the data that we have.

Q. Can science then identify the source of design at this point?

A. No, not at this point.

Q. Does intelligent design rule out a natural explanation for the design found in nature?

A. No, it does not rule it out.

Q. Could you explain?

A. Yes. Again, harkening back to the Big Bang theory, the Big Bang theory was proposed, and the cause of the Big Bang was utterly unknown. It's still utterly unknown. But nonetheless, the Big Bang theory is a scientific theory.

The Big Bang theory does not postulate that the Big Bang was a supernatural act. Although, you know, it simply posits no explanation whatsoever. In the same sense, intelligent design is a scientific theory advanced to offer -- advanced to explain the physical, observable facts about nature.

It cannot explain the source of the design and just leaves it as an open question.

Q. We've heard testimony about methodological naturalism. Are you familiar with that term?

A. Yes, I am.

Q. I believe you indicated in your deposition that you thought it hobbles or even constrains intelligent design, is that correct?

A. Yes, that's right.

Q. How does it do so?

A. Well, any constraint on what conclusion science can come to hobbles all of science. Science should be an open, no-holds-barred struggle to obtain the truth about nature. When you start putting constraints on science, science suffers.

Yesterday, I discussed a man named Walter Nernst who said that the timelessness of nature, the infinity of time was a necessary constraint on a scientific theory. Science had to operate within that framework. If he had prevailed, progress, real progress in science would have been severely constrained.

Another reason why methodological naturalism can be a constraint on science is because oftentimes people don't think -- don't separate neatly categories in their own minds. For example, I showed the -- I showed the quotation from John Maddox, the editor of Nature, who found the Big Bang theory philosophically unacceptable and was reluctant to embrace it because of that.

There are other scientists in the past, one named Fred Hoyle, who rejected the Big Bang theory because he did not like its non-scientific, extra-scientific implications. So to the extent that people confuse a scientific theory with extra-scientific implications that some people might draw from it, then that might -- that might be a constraint upon the theory.

Q. Despite these constraints, does intelligent design still fit within the framework of methodological naturalism?

A. Yes. Despite the constraints, it certainly does, just as the Big Bang theory does.

Q. Now we've heard some testimony about space aliens and time traveling biologists. And I believe you made some similar reference to that in your book, Darwin's Black Box, is that correct?

A. Yes.

Q. And why was that?

A. Well, this was, you know, a tongue-in-cheek effort to show people that, you know, intelligent design does not exclude natural explanations, although some, you know, explanations we might wave our hands to think up right now might strike many people as implausible, they are not, you know, utterly illogical.

And it was kind of a placemaker to say that maybe some explanation will occur to us or be found in the future which will, in fact, be a completely natural one.

Q. Now the space alien claim in particular seems to fall hard on the ear of a lay person. But has that been a claim that has been advanced by a notable scientist to explain the natural phenomena?

A. Yes, that's right. Surprisingly, in the year 1973, a man named Francis Crick, the eminent Nobel laureate who discovered the double helicle shape of DNA with James Watson, he published, with a co-author named Leslie Orgle, he published a paper entitled Directed Panspermia, which appeared in the science journal Icarus.

And the gist of the paper was that the problems trying to think of an unintelligent origin of life on Earth were so severe that perhaps we should consider the possibility that space aliens in the distant past sent a rocket ship to the Earth filled with spores to seed life on the early Earth.

Q. This was a claim advanced by a Nobel laureate?

A. Yes, Francis Crick.

Q. And the article in which his arguments appear, was this a peer reviewed science journal?

A. Yes, the journal Icarus.

Q. Was this just a tongue-in-cheek, so to speak, explanation on behalf of Francis Crick?

A. No, it wasn't. He mentioned it first in that 1973 article, and he repeated the same claim in a book he published in '88 and interviews he gave later on. And from what I understand, he still thought it was a reasonable idea up until his death recently.

Q. Sir, I'd ask you to direct your attention to the exhibit binder that I have provided for you, and if you could go to tab 14. There is an exhibit marked as Defendants' Exhibit 203-E as echo. Is that the article from Francis Crick that you've been testifying about?

A. Yes, this is Francis Crick's article on Directed Panspermia.

Q. Is the search for intelligence causes a scientific exploration?

A. Yes, it is.

Q. Again, do you have any examples that we could point to?

A. Well, one good example is one that I mentioned earlier, which is this project called the SETI project, S-E-T-I, which stands for search for extraterrestrial intelligence, where scientists use instruments to scan space in the hope of finding transmissions or some signals that may have been sent by extraterrestrial sources.

And they are confident that they could be able to distinguish those signals from the background noise, background radiation, electromagnetic phenomena of space.

Q. Again, that's a scientific exploration?

A. Yes, a number of scientists are involved in that.

MR. MUISE: Your Honor, I'm just -- do you intend to go to 12:30?

THE COURT: I was thinking more 12:15, unless you think that this is an appropriate break point. Your call.

MR. MUISE: I certainly have more than 15 minutes. This next section might be divided in that 15, so my preference would be to take the lunch break and come back and then complete the direct during the first session after lunch.

THE COURT: All right. We'll return then at, let's say, 1:25, this afternoon, after a suitable lunch break, and we'll pick up with your next topic on direct at that time. We'll be in recess.

(Whereupon, a lunch recess was taken at 12:04 p.m.)