Michael Behe's Amazon Blog

Apparently Behe is in a race to the bottom of the barrel in his latest "response".

In this response he attempts to address the crippling of the central point in his book "Edge of Evolution". His claim is that for malaria to develop resistance to CQ it must have two simultaneous mutations occur. Behe then took the probability of two specific mutations occurring simultaneously and tried to use it as a false "limit" of evolution. This is flat out wrong as clearly stated by the published literature. The literature clearly states that CQ resistance occurred gradually and that only having one gene mutation sill offers some resistance(in other words only having the one mutation is still beneficial instead of detrimental). The odds of two mutations occurring at the same time is irrelevant to CQ resistance in malaria and Behe knows it. Yet he keeps going back to it as if it does. Behe states:

"I stress in Chapter 3 that in the case of malarial resistance to chloroquine, multiple necessary mutations did happen in the membrane protein PfCRT. I also of course emphasize that it took a huge population size, one that would not be available to larger organisms."

He also quote mined Carroll in a vain attempt to support himself. He quoted Carroll saying:

"Behe's chief error is minimizing the power of natural selection to act cumulatively... Behe states correctly [my emphasis] that in most species two adaptive mutations occurring instantaneously at two specific sites in one gene are very unlikely and that functional changes in proteins often involve two or more sites."

Let me fill in what Behe cut out of that quote: "...as traits or molecules evolve stepwise from one state to another via intermediates."

Well, that changes things a bit. Behe is ignoring the fact that CQ resistance most likely occurred in multiple stages and then he had the audacity to take Carroll's quote out of context to give it the impression that Carroll was agreeing with his fictitious probability. The simply fact is that, yes, two mutations occurring at the same time IS very rare, but that has no bearing on Behe's claims regarding CQ resistance in malaria since it did NOT require two simultaneous mutations. In breathtaking fashion, after ignoring cumulative gene mutation in CQ resistance and then quote mining Carroll, Behe then admits that mutations CAN occur in sequence and accumulate, but then he tries to dismiss it by saying, "it is a non sequitur to leap to the conclusion that all biological features therefore can gradually accumulate". He tries to justify this by confusing his readers about beneficial and detrimental mutations. For one thing, natural selection filters out the detrimental mutations. Another more important fact is that the published literature clearly states that one gene mutation involved in CQ resistance imparts some resistance, and that resistance is ENHANCED by adding the second. The published works on CQ resistance are clear on this topic, CQ resistance most likely occurred gradually and the individual gene mutations ARE clearly beneficial. To watch Behe going through such contortions to try and defend his falsehoods is almost painful to watch.

Behe next attempts to address Carroll's demolishing of his claims about protein binding sites. Essentially Behe tried arguing in his book that binding sites are just too complex for evolution to account for. Carroll points out that this claim of Behe's rests solely on Behe's unfounded requirements for protein interaction. Carroll even ran through some simple math in his review that shows just how wrong Behe is:

"Very simple calculations indicate how easily such motifs evolve at random. If one assumes an average length of 400 amino acids for proteins and equal abundance of all amino acids, any given two-amino acid motif is likely to occur at random in every protein in a cell. (There are 399 dipeptide motifs in a 400-amino acid protein and 20 mult 20 = 400 possible dipeptide motifs.) Any specific three-amino acid motif will occur once at random in every 20 proteins and any four-amino acid motif will occur once in every 400 proteins. That means that, without any new mutations or natural selection, many sequences that are identical or close matches to many interaction motifs already exist. New motifs can arise readily at random, and any weak interaction can easily evolve, via random mutation and natural selection, to become a strong interaction (9). Furthermore, any pair of interacting proteins can readily recruit a third protein, and so forth, to form larger complexes. Indeed, it has been demonstrated that new protein interactions (10) and protein networks (11) can evolve fairly rapidly and are thus well within the limits of evolution."

Behe's only response to this is to misrepresent the cited references then to accuse Carroll of "begging the question" just as he did Jerry Coyne in his earlier "response". Behe does nothing to address the fact that his assertion in regards to protein binding sites is fundamentally WRONG.

I was still willing to give Behe the benefit of the doubt about the errors in his book. I was willing to entertain the idea that maybe he just got in over his head and didn't understand the subjects he was talking about. The naked dishonesty Behe has demonstrated in this "response" has cinched my opinion, the "errors" in Behe's book were deliberate.

You can read Carroll's review here:
http://www.sciencemag.org/cgi/content/full/316/5830/1427

References:

CQ resistance
http://linkinghub.elsevier.com/retrieve/pii/S009286740080447X

paper that Behe misrepresented
http://www.ncbi.nlm.nih.gov/sites/entrez?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=16172400

Your claim to objectivity, Mr. / Mrs Allen, is hard to believe, reading some of your other postings. But all right, let's take your arguments at face value. Here, it seems to me, you're missing the ball. You may even be swinging for the wrong fence.

Behe is not primarily making an apriori argument in this book. He is arguing from the actual history of evolution among pathogens. The point about chloroquine resistence is that in fact it has not arisen that often. Resistence to atovaquone, by contrast, appears to arise very quickly; every third person, he says. (59) He ascribes the difference to the fact that the former requires two mutations, the latter, just one.

If you want to refute Behe on this point, what you need to prove is that (1) CQ resistance actually demands more than two mutations. (To show that potentially profitable mutations arise more frequently than Behe claims), or (2) CQ resistance actually appears far more often -- exponentially more often -- than Behe claims.

If you can do one of these two things, then it seems to me that you will have made his position in this book less comfortable. Until then, you're just batting theory around. In fact, some of your arguments seem to actually make his point stronger.

David,
I'm not sure whether you simply haven't understood any of the critiques of Behe's probability estimates or if you are being willfully ignorant. I THINK you are just jumping in without any idea of what you are talking about in an attempt to defend Behe. For starters you say:

"If you want to refute Behe on this point, what you need to prove is that (1) CQ resistance actually demands more than two mutations."

Which leads me to think you don't understand what everybody is talking about in regards to CQ resistance. Behe's false assertion is that CQ resistance requires two SIMULTANEOUS mutations to occur. The reality is that the published literature clearly shows that the mutations for CQ resistance occur gradually, one mutation at a time. No one has said anything about CQ resistance needing more then 2 mutations. For one thing, that would be HELPING Behe's claim, not refuting it. As such, your claim that in order to "refute Behe" I would have to show that CQ resistance requires more then 2 mutations makes no sense. The whole point of this little exercise is to point out that Behe's false assertion greatly exaggerates the difficulty in CQ resistance by claiming that both mutations have to happen simultaneously. This one simple distortion allows Behe to radically alter the probability of CQ resistance arising via evolution(exponentially so actually). You seem to have totally missed the argument here. BOTH sides of the argument actually. Your entire comment is based around your faulty understanding of this issue.

Two other points: first you don't have to take my points at face value. I provided links to the original review that Behe was responding to as well as links to the published scientific literature that disproves Behe's claims about CQ resistance. If you don't believe me, read them yourself. Second, it's Mr. Allen. ;-)

In his first book, Behe argues that extremely complex biological structures could not have evolved by natural selection, because several interlocking parts, which are not beneficial without each other, could not be favored by natural selection.

In his latest book, Behe presents the argument from a different angle, looking at very simple evolutionary adaptations that require only a handful of mutations. Since the mutations are so simple, it is easier to determine if each mutational step is beneficial.

I am glad that S. Allen, David Marshall and Sean Carroll are addressing the facts, and taking the time to refer to published papers, rather than resorting simply to name calling.

Behe asks what if an evolutionary pathway of small, beneficial mutations does not exist for certain biological structures? What if a structure requires several interlocking parts to be beneficial? Parts which are not beneficial on their own? Once the parts are assembled and working, natural selection can improve the structure. But even if evolution uses existing parts as building blocks, how does the new structure get assembled in the first place? What if, by the very nature of a certain structure, it is impossible for a pathway of beneficial mutations to exist?

In his first book, Behe discusses very complex biological systems that he believes could not have been built by natural selection. It has rightly been pointed out that it is unfair to expect a complete reply because not enough is known about these structures and their history for a detailed evolutionary pathway to be developed. And besides, they are so complex that, as a practical matter, it would take too long to come up with a possible mutation by mutation pathway. So in his latest book, Behe deals with simple evolutionary adaptations that require only a handful of mutations.

The central point is whether all complex biological systems can arise as a result of small, step wise mutations, each of which is beneficial, or at least not very harmful. If a mutation is beneficial, then natural selection can favor it and it will spread in a population. Behe agrees that some individual mutations are beneficial.

However, his argument is that not every complex biological system could have arisen due to a sequence of beneficial mutations. What about a case where 10 mutations are needed before there is a benefit? If each mutation by itself is neutral, natural selection has nothing to act on. Then the probability of all ten specific mutations ending up in one organism, even if they are acquired sequentially over many generations, is vanishingly small.

Once a structure already exists, natural selection can fine tune it. However, Behe argues that in some cases natural selection is not sufficient, because multiple mutations are required, which are not beneficial in themselves. They are only beneficial after the basic structure is completed and functioning.

This is related to the argument about micro evolution versus macro evolution. Behe acknowledges that small mutations happen, which cause changes within a species. However, if his point about the lack of beneficial evolutionary pathways to form complex structures is correct, then natural selection cannot have been responsible for the huge differences between the major groups of living things with their vastly different body structures.

If materialistic evolution is the only cause of the diversity of life, then pathways must exist where multiple mutations are each beneficial in themselves. Until recent advances in DNA research and biochemistry, it has not been possible to propose a detailed, step by step, beneficial pathway to a new biological system. Today it is finally possible to test the theory that pathways of beneficial mutations exist in non-trivial biological situations.

In Behe's first book, he examines extremely complex systems to highlight the statistical impossibility of these structures arising by chance. This would be the situation if each mutation is not individually beneficial and thus favored by natural selection.

Some systems, such as the human blood clotting mechanism, have turned out to have reasonable pathways of development, possibly showing Behe was wrong in this particular case. These have been proposed as a result of studying simpler blood clotting mechanism in other animals. A complete, point mutation by point mutation pathway, each individually beneficial, has not been found, but it is reasonable that one may exist. Behe doesn't come right out and acknowledge this, but he does mention that he agrees that the "antifreeze" in fish blood evolved.

The other, even more complicated structure in Behe's first book, is the famous flagellum. The flagellum has a highly sophisticated motor which controls a long fiber propelling the creature it is attached to. Evolutionary biologists have pointed out that, since so much about the flagellum is not yet understood, it is currently unfair to expect a detailed evolutionary pathway to be developed.

Actually, as Sean Carroll lists in his end notes, N. J. Matzke and his associates have published several papers on possible pathways, which are much more detailed than the vague guesses that are usually proposed. In fact, his papers could be models for how evolutionary biologists should respond to the Intelligent Design argument. One of his early papers is reproduced at the following URL, with links to his more recent work:

http://www.talkdesign.org/faqs/flagellum.html#table3

However, as Matzke states, there are necessarily big gaps in his proposals, since so much is still unknown. But he believes that with further research, pathways could be found.

I disagree with Matzke and agree with Behe, that in principle, a pathway of beneficial mutations cannot exist for the flagellum.

Evolutionary biologists have claimed that the flagellum could have developed from previous structures and the modification of existing proteins.

Since we can plan ahead, it is easy for us to imagine taking existing objects and modifying them to make something new. But natural selection cannot plan ahead. Each mutation needs to be beneficial in itself. So how could each of the proteins that make up the motor of the flagellum have benefited the organism, when only a few of them had been attached together?

In his latest book, Behe looks at the argument from another direction, that of simple evolutionary adaptations that require only a handful of mutations.

While agreeing that in some cases a pathway of beneficial mutations exists for natural selection to work on, he argues that there are cases where such a pathway does not exist.

I am not able to evaluate whether Behe is wrong about the incremental benefits of some of the mutations he discusses. If in fact he is wrong, perhaps he will include a correction in the next edition of his book and look for other mutations that are not individually beneficial.

Behe mentions that some of the beneficial mutations that have been studied work by breaking something. That is, they interfere with the proper working of an existing system. It would be interesting to survey the literature to see how many mutations have been studied that add new functionality. That is, mutations that build up, rather than tearing down.

As an interesting side note, the New York Times has published several articles about evolution this week. A recent article in the New York Times mentions mutations that are harmful on their own, but are beneficial if certain other mutations have happened first:
http://www.nytimes.com/2007/06/26/science/26lab.html

In any case, it seems reasonable that Behe's estimates of the small probability of beneficial mutations is accurate. It makes sense that the time it takes an organism to develop resistance to a new drug is a good estimate of the mutation rate. Especially if resistance to a drug that requires only a single mutation occurs faster than resistance that requires multiple mutations.

I think the point that David Marshall was trying to make is that Behe has estimated a mutation rate that is much slower than has been generally accepted. If it takes decades for a pathogen to develope a two mutation evolutionary modification, then the mutation rate would seem to be too slow to explain the diversity of life only by mutation and natural selection.

If in fact, the pathogen actually develops four or sixteen mutations in the same number of decades, instead of only two, then the estimated mutation rate will be much faster.

If one believes that materialistic evolution is the only cause of the diversity of life, evolution must be true, almost by definition. If you assume natural, materialistic evolution is all there is, then natural selection acting on mutations, despite its limitations, must be the main explanation for all of life, because no one has been able to come up with something better. If mutations with natural selection are the main factors in the diversity of life, then every biological structure must have been formed by the accumulation of numerous small mutations, each of which was beneficial. But what if a conclusive example could be found of a biological structure where no combination of mutations were beneficial in themselves?

[Deleted by Amazon on Jun 27, 2007 2:17 PM PDT]

"...no protein binding sites - neither short linear peptide motifs nor any other - developed in a hundred billion billion (1020) malarial cells. Or in HIV."
Excuse me, but what? Can you please clarify this sentence, please, in the context of HIV?

Joseph,
Irreducible Complexity is nothing more then a god of the gaps argument. So far every system that Behe has identified and claimed was irreducible has been shown to be reducible. His claims about the flagellum were proven wrong(cut it down by 40 components and you have the type III secretory system) nearly a decade ago. His claims that the blood clotting cascade were irreducibly complex were shown to be false when it was pointed out that the dolphin's immune system worked just fine without a component that Behe said was required. His latest claim of an irreducible system is contained within this book where he argues that cilia HAVE to have the IFT to form. This is again shown to not be the case by the published literature.

As you can see irreducible complexity is nothing but a god of the gaps explanation. When one system that was proclaimed to be IC is explained or demonstrated to NOT be IC, then Behe simply picks another system and says, "What about that one?".

Is the "intelligent designer" irreducibly complex? If so, who designed him, ad infinitum? But if not, then did the intelligent designer himself evolve?

Either way, ID loses.

Mr. Behe,
how does your probability calculation deal with the documented fact that a chloroquine-resistant species of malaria without one of the mutations you discuss (at position 220)? Isn't this an example of sequential evolution?

Secondly, are you serious when you claim that specific protein-protein interactions evolve only with difficulty? If so, you are making a good case for revocation of your biochemistry diploma.

Pretty much every protein will interact with any other protein, depending on conditions and concentrations (as anyone who has performed pulldown or shift assays can tell you; but you should know that yourself by now). If a weak interaction provides a selective advantage, further mutations ANYWHERE in the interaction region that strengthen this interaction IN ANY WAY will be highly selected for. This does not require specific point changes at specific positions, but any kind of change at any position where initial weak binding contact occurs; and this counts for interaction surfaces of BOTH proteins (i.e. both can change).

Given the scope of freedom most proteins have to change sequence without impairing function, I find your "argument" so removed from reality that I can't believe anyone who has taken a junior biochemistry class would buy it.

This, incidentally, has nothing to do with the reality of evolution. Even if theory of evolution was incorrect, and there really is a Designer, it makes no difference. Based on hard, published data about known behavior of proteins and known incidence and effect of mutations, your arguments are completely incorrect.

Dear Mr. Kwok,

Thank you for taking the time to reply.

Yes, there are a number of eloquent writers that have been published in the field of evolution. I enjoy reading Dawkins, even though I don't agree with him. I wish Mayr had our current knowledge of DNA so that he could have come up with a much more rigorous definition of "species."

The study of Intelligent Design is still a very young field, even younger than micro biology and the study of DNA. At this early stage, there is not enough data to make statements about what a designer might be like, or how design was imparted to life on earth (if it was). And, unfortunately for Intelligent Design, as well as for Cosmology and Evolution, we can't create a new planet and wait around for several billion years to see how life develops.

If I may offer an analogy from physics, this is similar to someone asking Copernicus for a detailed explanation of gravity, 100 years before Newton, when he first proposed that the sun and not the earth is the center of our solar system. Just as Copernicus believed that a sun centered system fits the data better, but couldn't explain the details before Newton invented calculus, there are some who believe that since life shows signs of being designed, perhaps there was a designer. But the details of how and who are beyond our current level of knowledge.

But, since you mentioned it, let me engage in some very non-scientific speculation. Please do not evaluate Intelligent Design on the basis of my personal, unscientific speculations!

Perhaps every 100 million years or so, at the time of the development of each major new form of life, someone from an interstellar bio-tech firm injected designer DNA into some cells on Planet Earth, giving the instructions for the new body plan. After that, the organisms could continue to diversify by the normal processes of evolution.

Or, perhaps the first life was designed with the complete DNA instructions to make every kind of living thing that exists on earth. As environmental conditions changed, different Prue-existing genes were expressed resulting in fish, reptiles, mammals, etc. Over millions of years, fish lost unneeded genes that were specific to mammals and mammals lost genes that were specific to fish, etc. On a much smaller scale this would be analogous to stem cells, which initially have the capability to develop into various different cell types, but eventually specialize. As has been pointed out, based on current observations, it appears that there would be too many mutations in parts of DNA that were not being used, but would be needed in the future, for this hypothesis to be valid. But perhaps when life first was formed, the mutation rate was different.

What if common descent looks like it happened, but actually the designer separately constructed several different models, some using similar designs and parts, such as DNA, five fingered hands, etc?

When I referred to "natural, materialistic evolution", I meant in contrast to something synthetic, or non-natural. The contrast would be something that was designed by an intelligence, like humans, rather than something that resulted from just the forces of matter and energy in nature. Another, more extreme contrast to materialistic evolution would be theistic evolution, which proposes a supernatural rather than natural, materialistic origin of life.

I am also interested in string theory. Although it has been referred to as "bad science" because there are currently no experiments that we can perform to prove or disprove string theory. Perhaps when the new super collider is finished in Europe clues about string theory will be found. When I met Brian Greene earlier this year, he suggested several future experimental confirmations of string theory.

We are living in very exciting times, because we are getting to the point in our knowledge of biochemistry and the study of DNA to be able to look for direct evidence for or against the formation of biological structures by the process of mutations favored by natural selection.

Regards,

Joseph Fischer

Mr. Allen (glad that's cleared up): It's you who missed my point; but maybe I didn't make it clear enough. I know you've challenged Behe's assumption that the two mutations need to occur at the same time. Maybe so; I had my doubts when I read the original claim, and I do think that point needs to be addressed more clearly.

But Behe's argument is not primarily based on theoretical calculations about how difficult mutations might be. They are based in historical data about how difficult they actually have proven to be. Given a certain number of malaria patients administered a certain drug, given a number of bugs per patient, it should not be difficult to calculate how often malaria does in fact develop resistence to a given drug. That is Behe's primary argument.

If, as I think you claim, the first mutation confers a slightly positive effect, and is therefore selected for, it would be even more puzzling why the appearance of both mutations together is so rare.

That's the point. Again, I suspend judgement as to whether Behe is right or not; but it looks to me like you're hunting the wrong fox.

My point about whether two or more mutations would be needed was not to question the fact that it does take two mutations, but simply to point out a logically possible response to Behe, whether it is empirically viable or not.

Joseph,

ID is NOT a new field of science. ID is not science. It has no scientific theory, it does not even have a scientific definition. It makes no predictions. It is in essence indistinguishable from nothing. ID most certainly is not comparable to any of the major scientific theories of history.

The truth behind ID is that it is nothing more then dressed up creationism designed to try and bypass the courts so it can sneak into high school science classes. It failed at that task it's first time at bat. As such ID is pretty much useless to the creationist movement now. It is little more then a buzz word that is used for DI propaganda now.

[Deleted by Amazon on Jun 27, 2007 7:15 PM PDT]

"I know you've challenged Behe's assumption that the two mutations need to occur at the same time."

No, I didn't challenge his claim, the peer reviewed and published science did. I posted a link to the relevant research which clearly shows Behe is wrong.

"But Behe's argument is not primarily based on theoretical calculations about how difficult mutations might be"

Actually, yes it is. It's the core argument in his book which is clearly evident by the number of times he references his bogus probability of 1x10^20. Attempting to define false "limits" to evolution was the entire purpose of his book. Behe's false probability argument was the best he could come up with for this.

I not even sure what you are trying to say when you are talking about "given a certain number of bugs per patient" etc.

"If, as I think you claim, the first mutation confers a slightly positive effect, and is therefore selected for, it would be even more puzzling why the appearance of both mutations together is so rare."

You are correct. I don't think you intended to do it, but you just saw through Behe's argument. The simple fact is it's no where near as hard as Behe leads his readers to believe for sequential mutations to both occur and combine. It seems you are starting to understand. :-)

"My point about whether two or more mutations would be needed was not to question the fact that it does take two mutations, but simply to point out a logically possible response to Behe, whether it is empirically viable or not."

David, I don't think you understood what was being discussed. Otherwise you wouldn't have said that in order to "disprove" Behe I would have to show that it takes more then 2 mutations for CQ resistance to arise. Again if you don't want to believe me, go back and use the link I provided and read it for yourself. The cited article lays it out in black and white.

[Deleted by Amazon on Jun 27, 2007 7:00 PM PDT]

Dear Mr. Allen,

Thank you for your comments.

In a "god of the gaps" situation, everything we don't understand is attributed to the actions of a god. For example, if we don't understand how a certain DNA sequence came to be, we could say that a god inserted it there. Then, when we find a natural, biological explanation, we understand the situation and no longer attribute it to a god. The damage to science caused by a god of the gaps, is that it can tend to slow the rate of scientific inquiry. If we assume that something happened because a god did it, then there is little motivation to study it further. After all, who can understand the ways of a god?

But this has not happened. More research is taking place in biology than ever before. And besides, Intelligent Design hypothesizes an intelligence, but doesn't get more specific. The design could come from an advanced civilization from another solar system.

I understand what you are saying about systems that were said to be irreducibly complex, turning out to be reducible. In fact, in one of my posts above, I agree with you that Behe probably was mistaken when he called the human blood clotting cascade irreducibly complex.

In the other examples, I think the problem is with the wording of Behe's definitions. I think the underlying concept is still valid.

Like any good scientist, Behe should modify his theory to include the situations that have been pointed out to him. Behe should modify his definition of Irreducibly Complex, if he hasn't already. The new definition of a system could allow for parts that are themselves beneficial and have been improved by natural selection. The definition should acknowledge that we don't always know if a particular biological system would still function if a part or two were removed. So although by Behe's original definition, many, if not all of his examples turned out not to be Irreducibly Complex, in many cases the systems could become irreducibly complex by removing a few parts that are not strictly required.

If you start removing parts, in many cases, at some point you will reach a situation where the system will no longer perform its current function. In our example of the flagellum, if you start removing the 40 parts that you mention, at some point the flagellum will stop working. However, this will occur well before the structure has been reduced to a type III secretory system and starts functioning as one. The problem is that "in-between" stage where the structure has too many parts to work as a secretory system, but not enough parts to work as a flagellum. How could a non-working structure be beneficial enough to an organism to be favored by natural selection? And how could the mutations that add each of those 40 parts benefit the system each step of the way, so that they will be favored by natural selection? Until most of the parts are added, the system won't function as a flagellum. The traditional explanation is that the in between steps produced systems that performed other functions that benefited the system. Yet it seems unreasonable that there would be all those different systems, with functions that each offered a selective advantage to the organism and happened to lead up to the formation of the flagellum.

The case of the cilia is similar. Yes, Behe seems to have been misinformed about the need for IFT. Yet his basic, underlying concept still holds. If you remove enough unnecessary parts, eventually you will reach the point where the system no longer functions as a cilia. At this stage, the cilia is Irreducibly Complex. Just as an example, say that the cilia also requires 40 parts. Again, how does a system, which doesn't have enough parts to work as a cilia, benefit the organism enough to be favored by natural selection? And how can natural selection favor the step by step assembly of each of those 40 parts during that stage when there are not yet enough parts for the system to function as a cilia? This is the basic idea of Irreducible Complexity theory.

Regards,

Joseph Fischer

"But this has not happened. More research is taking place in biology than ever before. And besides, Intelligent Design hypothesizes an intelligence, but doesn't get more specific. The design could come from an advanced civilization from another solar system."

Researh IS taking place in biology, however there is none taking place in ID. Without a scientific framework to start from, there simply is no way to do research into ID. Not identifying who they think the designer is has nothing to do with that fact. The creationists at the DI intentionally avoid discussing who the think the "designer" is for the simple reason that they are trying to trick people into thinking that ID is separate from their religion. They would have a hard time of that if they revealed their designer to be the god of the old testament.

"In the other examples, I think the problem is with the wording of Behe's definitions. I think the underlying concept is still valid."

It's not his definition, it's his hypothesis. The evidence clearly shows his hypothesis of IC is incorrect, he has simply chosen to ignore the evidence and keep asserting IC as if it were valid.

"Like any good scientist, Behe should modify his theory to include the situations that have been pointed out to him. Behe should modify his definition of Irreducibly Complex, if he hasn't already. The new definition of a system could allow for parts that are themselves beneficial and have been improved by natural selection."

The system you just described already has a name, it's called evolution. ;-)

Oh, and one little quibble, I wouldn't call Behe a good scientist. For one he doesn't do any research, and two the errors he has made in this book indicate he either has no understanding of basic evolutionary biology or he is being intentionally dishonest.

"If you start removing parts, in many cases, at some point you will reach a situation where the system will no longer perform its current function. In our example of the flagellum, if you start removing the 40 parts that you mention, at some point the flagellum will stop working."

You fail to understand what Behe is trying to describe with his failed hypothesis of IC. Behe's conjecture is that in an IC system all the parts HAVE to be there or the entire structure is useless. Essentially he uses IC to argue that the structure could not have evolved because all the pieces have to be present at the same time or it's all useless. This is clearly false. What Behe overlooks is that the FUNCTION of these various pieces can change. That's what you see if you look at the type III secretory system. It's composed of about 40 of the same pieces as the flagellum, but it has a completely different function. This demonstrates how supposedly IC systems can be reduced to smaller components performing different jobs. Once you realize that these supposedly rigid IC systems are actually very versatile in their function, it becomes clear how easily evolution could produce them in a step wise fashion.

"The case of the cilia is similar. Yes, Behe seems to have been misinformed about the need for IFT. Yet his basic, underlying concept still holds. If you remove enough unnecessary parts, eventually you will reach the point where the system no longer functions as a cilia."

What you are describing is the classic god of the gaps scenario. Essentially what you are doing is that after a supposedly IC system is shown to have a reducible component, you are then pointing to a smaller component and saying, "well, then THAT'S IC". What you are doing is shoving IC into ever smaller gaps as the larger gaps are filled in.

"Again, how does a system, which doesn't have enough parts to work as a cilia, benefit the organism enough to be favored by natural selection?"

Because the smaller precursors to the cilia performed OTHER functions that were useful before being incorporated into the cilia itself. This is where Behe tries to confuse his reader. He has them convinced that all of these various parts are only useful for performing ONE function in regards to the cilia. He glosses over the fact that in a different structure they may have a different function.

[Deleted by Amazon on Jun 28, 2007 9:13 AM PDT]

"If, as I think you claim, the first mutation confers a slightly positive effect, and is therefore selected for, it would be even more puzzling why the appearance of both mutations together is so rare."

In addition to the other responses, let me add another complicating factor to what may seem like a fairly straightforward calculation.

Prior to the evolution of the modern resistant strains, malaria parasite was quite sensitive to chloroquine, and (IIRC) the doses given to people far exceeded the minimum lethal dose for the parasite. This means that partial resistance to the chloroquine could have evolved many times, but was not sufficient for survival.

The modern strain thus required a less-probable mutation, or (more likely in theory; I am not conversant enough with the epidemiological data in this case to know whether it is so in practice) an opportunity for stepwise mutation towards full resistance through people who did not receive a sufficient dose (old medication? noncompliance?).

In any case, we are looking at a mutation that has to overcome a very high initial boundary, failure at which leads to death and immediate elimination from the selection.

In nature, evolution rarely deals with challenges of this kind. Instead, we have either slight negative selection (failure to overcome the hurdle leads to a slight to moderate impairment, but does not eliminate the individual), or slight positive selection (there is no hurdle; a mutation simply confers a small or moderate positive advantage to the creature).

In both cases, a mutation that is only a small step towards what the organism finally ends up with allows survival and procreation. In malaria, any mutation that does not meet the minimum requirement (level of resistance that allows for survival) means elimination without progeny, making evolution a lot harder.

Prevention of such scenarios in bacteria is one of the main reasons doctors and pharmacists will always urge you to finish the whole course of antibiotics, even if you feel completely cured midway through it. If you don't, a few partially-resistant bacteria may survive the shortened/reduced exposure (where the full dose/length would kill them), and then serve as a platform for development of fully resistant strains.

But that is a whole separate topic. In any case, this is yet another reason why Behe's analogy doesn't work (besides hard evidence to the contrary of his propositions).

Response to Kalla, John Kwok & S. Allen:

Kalla, you have a good insight about the differences between evolution in the presence of fatal poisons and more normal situations where slightly beneficial mutations are allowed to accumulate.

John Kwok, I look forward to reading the reviews by Miller and Dawkins. Dawkins is a wonderful writer and Miller has contributed reasoned and informed discussions of intelligent design.

S. Allen, I like your reply to my suggestion about modifying Irreducible Complexity. I suggested that Behe change his definition. You replied that:

"The system you just described already has a name, it's called evolution. ;-)"

As a result of further investigation, it may turn out that your statement is correct. However, I think we need to know more about the biochemical details of the step by step mutations, before we can make that determination.

It is clear from Kalla's comments and other remarks, that there is much more to be learned, even about a well studied, relatively simple organism, such as malaria. And you have all made some very valid points about mistakes in Behe's book.

However, it seems to me that it is still possible that the basic core of irreducible complexity is valid. Is there in fact, a pathway of step by step mutations, each of which is beneficial, that result in complex biological systems and structures? We all agree that natural selection can improve a structure that already exists. The fundamental question is whether natural selection can build a new, non-trivial structure.

Yes, S. Allen, the fact that a type III secretory system could be modified to become one of the building blocks of a flagellum is a key insight. Miller has written clearly about this and I am surprised that the ID camp hasn't responded to this more forcefully, such as asking where the high-speed, rotory motor came from.

But until the details are filled in, there is still the question of whether there is a step by step path from the secretory system to the flagellum. Since this is such a complex system, it is currently unfair to expect all the details of such a path to be written down.

So in his current book, Behe is considering much simpler systems. However, from Kalla's post and other comments, it looks like we don't yet know enough about even these simpler systems.

Did it take two mutations or four mutations? How many times did the mutation arise before it was able to spread world wide? Is mutation under the influence of toxins different than "normal" evolution?

The New York Times had an interesting article about intensive research being done on e. coli.

http://www.nytimes.com/2007/06/26/science/26lab.html?_r=1&ref=science&oref=slogin

Because they reproduce so quickly, scientists have been able to study thousands of generations. They have seen some surprising results. Unfortunately, so far, we don't know enough about the underlying DNA to understand how these changes are happening. Perhaps further research will show new structures evolving before our eyes. But until then, I remain unconvinced about the power of natural selection.

-Joseph Fischer