Home » Intelligent Design » The Flagellum Challenge for Darwinian Evolutionists

The Flagellum Challenge for Darwinian Evolutionists

When IDists hypothesize that the bacterial flagellum is irreducibly complex they explain that the hypothesis may be falsified by a detailed and plausible pathway whereby random mutation and natural selection could have built it up. We don’t ask that it be proven that’s how it happened only that it be demonstrated it can happen that way. Charles Darwin himself in The Origin of Species, chapter 6, anticipated this argument against his theory and acknowledged it would spell doom for his theory:

If it could be demonstrated that any complex organ existed, which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down. But I can find out no such case.

150 years later Darwin’s critics believe they have found a complex organ, the bacterial flagellum, which could not possibly have been formed via numerous, successive, slight modifications.

This is all well and good and certainly does not prove Darwin’s theory is wrong because, as even Darwin must have known, proving a negative is impossible. We can never, ever know that no Darwinian pathway is possible. All we can know is that no proposed pathway can explain it. Fortunately for IDists science doesn’t require proofs. According to our most widely accepted philosophy of science what science does require in cases like these is a method of falsification. In science one needn’t prove a negative if there exists a way to falsify the positive. This is what separates pseudo-scientific theories that explain everything, and thus explain nothing from real scientific theories.

This takes us back to the beginning where I explained that the hypothetical irreducible complexity of the flagellum may be falsified by showing any possible and plausible Darwinian pathway. The hypothesis that the flagellum is irreducibly complex is good science.

Now for the challenge. I had challenged a commenter here (Tiax) to explain to me a scientific method by which the theory that the bacterial flagellum evolved via random mutation plus natural selection could be falsified. My challenge was met with the sound of crickets chirping. Therefore I am putting this challenge in a more prominent position.

If no scientific method of falsfication can be provided then the so-called evolution of the flagellum is nothing but pseudo-science. My position is that Dembski’s design detection theory is indeed science and that it is the only falsification method by which hypothetical flagellum evolution can be rescued from the pseudo-science trashheap. If design detection isn’t science then neither is any theory of the flagellum evolving. Maybe Judge Jones needed to dismiss more than just design detection as “not science”, eh?

Who can provide for me a scientific method by which the theoretical Darwinian evolution of the bacterial flagellum can be falsified?

Fair warning: I’m not allowing any comments here that do not propose a scientific method of falsifying Darwinian evolution of the flagellum so don’t waste your time composing anything else.

I apologize for deleting comments that are in agreement. It’s understandable that you’d agree.

One commenter attacked the idea that falsification is widely accepted by philsophers of science as a demarcation of science and non-science. He may be right but my point is about what is legally science and what is not which is why I mentioned Judge Jones. Falsifiability is a legal test for what is science and what is not:

Falsifiability was one of the criteria used by Judge William Overton to determine that ‘creation science’ was not scientific and should not be taught in Arkansas public schools. It was also enshrined in United States law as part of the Daubert Standard set by the Supreme Court for whether scientific evidence is admissible in a jury trial.

I’ve had to delete a number of posts proposing ways the flagellum could have evolved. That’s not the challenge. The challenge is to give a method of falsification.

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

19 Responses to The Flagellum Challenge for Darwinian Evolutionists

  1. There was more to this comment but it was not relevant to the challenge and was removed. -ds

    Another simple method would be to conduct a search of bacterium and show that one or more could not have been derived from any existing or potential ancestral forms. Find one with a flagellum and a non-DNA/RNA coding process. You will have falsified common descent of life on this planet, which requires mutation and natural selection.

  2. 2

    If you could show that no variation of any of the genes coding for the flagellum would allow the bacteria to have a flagellum, then you’d go a goodly distance. Most genes seem to have the property that if you change an amino-acid here or there, you still get out an (at least mostly) working protein. If none of the proteins in the flagellum had this property, you’d be close.

    To clinch this, you’d then have to examine variants of every other protein, to show that they couldn’t substitute. How many variants? How many proteins? I don’t know; but I imagine if you covered everything with a conformation close to the original that would go another good distance to being convincing.

    I’m prevaricating on the exact amount of work needed as I’m by no means an expert in the field.

    You could also try Michael Behe’s suggestion of selecting a great many generations of bacteria for mobility to see what happened. The difficulty here is that you’re proposing a very long running experiment.

    There are also some excellently preserved strata of protoists in ice cores – there may also be some of bacteria. Were there a good strata going back many years through which one could see non-flagellum bearing bacteria transformed into flagellum-bearing bacteria by a non-natural selection process this would also be some evidence.

    If these suggestive evidences were put together, this would probably comprise a useful falsification. I’m tentative on account of the “extraordinary claims requiring extraordinary evidence” dictum.

    Also you could sit and observe bacteria. If you had some method of coaxing an/the Intelligent Designer to do his thing (prayer? broadcasting to the alien intelligences?), you’d see a non-motile strain getting spontaneously transformed (kazzam!) into a motile strain. That would definitely be convincing.

  3. Edited to include only the proposed falsification method. -ds

    Potential falsification of flagellar evolution could be based on the lack of plausible homologous structures.

  4. “If design detection isn’t science then neither is any theory of the flagellum evolving.”

    I’ll take this to mean that I can demonstrate that a given model of the flagellum’s evolution is falsifiable

    No. This only falsifies one possibility. It’s the same argument used against ID i.e. “Well if we show that the flagellum evolved you’ll just pick out something else and say that’s irreducibly complex”. That is a solid argument and what’s good fo the goose is good for the gander. Falsifying one instance leaves the general hypothesis intact. To get around this I chose DNA and ribosomes as what would falsify ID for me – if it can be demonstrated that those structures which are common to all known living things can evolve without intelligent input I’ll concede that everything else could have evolved without intelligent input. Something tells me you won’t be willing to concede that if Matzke’s proposal can be falsified you’ll concede that no Darwinian pathway is possible. -ds

  5. At some point I want to start a new thread where these proposed methods of falsification can be discussed. So please, if you have a method in mind please add it to the comments. I don’t really care what it is as long as it attempts to define a scientifically sound way to falsify the NeoDarwinian theory that mutation + selection is a mechanism sufficient to create a bacterial flagellum. No bloviating. I’ll give everyone a chance to bloviate on a separate thread. I’ve already put forth in the article that mathematical design detection (Dembski) is one proposed way so there’s no need to add that again.

  6. One quick point, and then some ways to falsify the standard model for flagellum evolution.

    First, I think it is important to remember that it is not just RM + NS that is important. One must also consider neutral mutations that fix by drift.

    Genetic drift in prokaryotes? Maybe you should think about that a little longer. -ds

    That being said:

    I think there are several classes for types of rejection. The first is direct observation. Unfortunately, for historical cases, direct evidence can not be obtained without a time machine. This problem plagues the subject of history as well and makes evolutionary biology a different science. But, in principle, if time travel were possible, the easiest test would depend on 1) going back in time and 2) collect samples of all bacteria. With this impossible data set, one would have to show that mutations appear to follow a poisson process that we expect and that whatever we define the final flagellum to be, we would have to show that version starting at frequency 1/N and reaching fixation. If we could show that this didn’t happen, we would reject.

    OK, not likely. The second type of rejection would be indirect evidence for non-Darwinian process. There are many possible types of nonDarwinian process. Under the umbrella of design, one might include alien manufacture. Thus, one kind of rejection would be to identify specific evidence for such a process. For example, if we found a platinum disc on the moon that contained video footage of aliens manufacturing the flagellum, this would likely reject. Basically, a Zapruder film.

    OK, also not likely. Finally, aside from design inference, there are other types of inferences. Darwinian evolution requires time. Thus, a Darwinian explanation would be rejected if one could soundly infer instantaneous (literally) proliferation of flagellum posessing bacteria. For example, suppose there is a specific byroduct of flagellum use (freon for air conditioners). If one could demonstrate that this byproduct appeared immediately at steady concentration, rather than ramping up gradually, this would reject Darwinian evolution. But when I say immediately, I mean immediately. In fact, faster than the growth of Bacteria itself would be possible.

    And finally, falsification would be possible if one could infer the exact genome sequence of the pre-flagellum bacteria, establish a date for it (based on rate assumptions) and show, conservatively, that the SHEAR number of mutations (forget about the combinatorial problem for now) required to go from point A (pre-flagellum) to point B (flagellum) in the estimated time T would be near impossible given the mutation rate (time 50, to be conservative).

  7. Another way any hypotheses of flagellum would be effectively falsified is if within the genome a signature of the designer was discovered, something unambiguous to all people of normal intelligence. If say, some sort of bitmap image was encoded within the genome similar to the illustrations we see on the Internet, with perhaps, some known human language caption underneath that said, “This is a designed mechanism. Stop arguing about it. Sincerely, the Designer(s)”, I think that would do it.

    Interesting. Since I asked for a scientific method of falsification I must assume that you either think “Designers” are scientific or you strayed outside the bounds of the question and gave me an answer that postulates a supernatural being. Which is it? -ds

  8. I think that for most scientists the argument that a biological structure is intelligently designed would be convincing if the designer could be identified scientifically. I see this as a major point of contention between the ID community and the rest of the scientific community. ID proponents say that it is not necessary to identify the designer, but the scientific community says that in order to propose intelligent design as a valid scientific theory the designer must be identified. Until one community or the other capitulates on this core philosophy there will always be arguments between the two.

    You’re saying the designer must be identified before we can even legally tell students in public school that it’s a possibility? If we were demanding that ID be taught in public school as a fact immune to criticism like NeoDarwinism is currently taught then you’d have a point. But all we’re asking is that the possibility of design be introduced. Just the possibility. Evidently that’s too much. We have to prove not just that some things might be intelligently designed but we have to be able to identify the designer. If the NDToE were treated the same way we wouldn’t be able mention the natural evolution of anything that wasn’t witnessed. Talk about a double standard. -ds

  9. “Interesting. Since I asked for a scientific method of falsification I must assume that you either think “Designers” are scientific or you strayed outside the bounds of the question and gave me an answer that postulates a supernatural being. Which is it? -ds”

    I think the former seems to fit squarely within domain of methodological naturalism, regardless of whether the designer(s) is/are natural or supernatural. It would certainly be a “scienfic event” to discover a purposefully embedded complex undeniable signature within a genome, or perhaps all genomes. The question of how it got there would undoubtably stir up quite a debate.

    Good answer. I was however looking for falsification methods other than design detection. The signature you describe would be a designed signature and the nature of it would preclude argument over whether or not it could possibly be generated by chance. I suspect both of us being engineers we know of the proclivity for our kind to cleverly tag our work. -ds

  10. 10

    You could also: 1. Show that mutations never occur in the genes which codify the flagellum, or they are error-checked to no allow mistakes. 2. Optionally show that mutations that do arise are either not selected out (if they are disadvantageous) or selected for (if they do convey an advantage). If either of these factors is not present, even in modern flagellum, then it would cast doubt on whether RM+NS was working on it in the past. This is a necessary falsification because one of the primary assumtions behind proposing RM+NS as the designing process is the assumption of uniformity. Showing that either or both of these points do occur gives greater confidence that RM+NS could have been at work in the past. This is a falsification which ID can not duplicate for itself.

  11. ds,
    In response to your reply to my last post, I don’t think it’s such a double standard. Natural selection as a mechanism for the creation of biological structures is based upon biological, chemical, and physical processes that have been observed in nature. Although the details of how these processes worked to create life and the biological structures we see today may be unclear, the general concept arises from processes redily observed. ID, on the other hand, is predicated on a process that has never been observed in nature to be responsible for the creation of life and biological diversity. Until that process (the designer) is observed in nature the scientific community will have to stick with observable phenomena and extrapolate from there.

    Intelligent design in nature has been observed in recent history. The question isn’t whether intelligent design is possible – we already know it is possible because humans have become able to manipulate genomes for their own purposes – but whether this is the first time in the history of the universe that an intelligence capable of genetic engineering has existed. Try again. -ds

  12. Let’s just assume for a moment that further information than is known (such as the 3 billion+ base pairs of the human genome) is spatially encoded into the structure of DNA and that there is a latent library of mechanisms. What if we’re running an experiment such as Behe has suggested and all of the sudden–BOOM!–we have a bacterial flagellum. Not gradually through a direct pathway but abruptly and in final form (and not by external intervention). I would think that such evidence would be more than enough evidence for most people to rule out RM+NS. On a side note, obviously such an event would provide great benefit in figuring out how the data for this hypothetical latent library is encoded.

    1. Show that mutations never occur in the genes which codify the flagellum, or they are error-checked to no allow mistakes.

    I doubt ANY data replication method is completely error proof. In E. coli, replication proceeds along DNA at a rate of about 1000 nucleotides per second and a wrong base is incorporated about once every 10^5 steps. But it contains proofreading activity which can catalyze the hydrolysis of the phosphoester bond. This proofreading activity itself makes an error about once every 100 hydrolyses. Together, this gives an error rate of about 10^7. And for reference, the 1 in 10 million error rate doesn’t take into consideration error correction by separate DNA repair enzymes, which result in an overall error rate of about 10^10. The human genome contains about 3.2×10^9 base pairs which means that on average one error is made during genome replication.

    For comparison, here are example “acceptable” error rates in human designed technology:
    1 in a million – Acceptable voice quality through a T1 channel
    1 in a million – Microwave signaling error rates
    1 in a billion – Modern hard disk error rates

  13. Was thinking about it more and while evidence of a bacterial flagellum coming from a bacterial flagellum would be great it still wouldn’t directly falsify indirect Darwinian pathways via RM+NS.

  14. Patrick, it’s like DaveScot said, “The challenge is to give a method of falsification” and not how it might have evolved. In other words, what might we expect to see if RM+NS was not responsible. I proposed that you might see that RM+NS were not operating on that particular structure, if they were not responsible for it. This is sufficient to falsify RM+NS as a valid proposal. Without an alternative proposition which can meet similar criteria (show evidence of an ID at work today to justify extrapolation into the past) evolution becomes the default designer of the flagellum, whether you’ve got an indirect pathway or not. Again, I’m not trying to show how it evolved, but that RM+NS can in fact be falsified as a theoretical process.

    Now, if you want to dig deeper, you could ask if the indirect pathway hypothesis is falsifiable. I’m sure you’re aware of the proposed falsification, find structures which only use some fraction of the final mechanism (TTSS?). The mere existence of functional partial systems is sufficient to establish indirect pathways as a falsifiable proposal. Beyond that, you’d have to identify a particular pathway that you wanted to falsify and determine what we should see if that pathway were incorrect. I don’t have a way to falsify all indirect pathways in general without doing an exhaustive search. Nevertheless, falsification is generally quite easy to establish for specific evolutionary methods because it is a simple matter to suggest results that would be seen if your hypothesis is false. What would we expect to see if ID was incorrect?

  15. while evidence of a bacterial flagellum coming from a bacterial flagellum

    Brain fart. Meant to say “coming from a latent library”…

  16. “ID proponents say that it is not necessary to identify the designer, but the scientific community says that in order to propose intelligent design as a valid scientific theory the designer must be identified.”

    When anthropologists discover ancient tools, do they need to know the identity of the designer in order to reasonably conclude the tool was designed?

    Please tell us what knowing the identity of the designer does for us with regards to detecting design?

  17. “Good answer. I was however looking for falsification methods other than design detection. The signature you describe would be a designed signature and the nature of it would preclude argument over whether or not it could possibly be generated by chance. I suspect both of us being engineers we know of the proclivity for our kind to cleverly tag our work. -ds”

    There are at least two scientifically valid questions that could be raised with regards to a given feature in a genome:

    1) Can feature X come about by NDE in principle, given what we know about nature?
    2) Did this instance of feature X come about by NDE or not?

    If you’re talking about falsifying the *possibility*, given what we know about nature, that a given feature of a genome could come about via NDE, then I would agree. I don’t think it is possible to falsify the possibility. NDE can “explain everything.” (Which means it explains nothing until a blow by blow account can be given for a presence of a paricular feature.) But if we’re talking about a particular instance of a feature having a signature within in the genome, where the designer takes credit, I would have to consider that this *given instance* of the feature having *actually* come about by NDE would stands as falsified. I realize this is not what you were driving at. :)

  18. I would like to add that NDE explains everything in general, because it explains nothing in particular. Impossible to falsify such a concept because it does not deal with particulars. Rather it is an ideological axiom. Axioms are not subject to falsification by definition. And this particular axiom is simply the mirror corollary of an a priori bias against design, supernatural OR natural.

    It’s amazing to me how bad philosophy can pass as a “scientific” paradigm.

  19. If the flagellum were to have developed via RM+NS, there must have been a precursor to the flagellum which requires one and only one mutation event.

    If we consider a real-world flagellum, and we subtract from our consideration all of the genes in that flagellum which do not disable the dear thing — the “reduceable” genes, then we have the minimal flagellum.

    Lets now factor in the view that precursor genes may have performed two functions, become copied, then each copy specialized to a single function. Let’s attempt to reduce the “minimal flagellum” by using this logic as far as the real genes can practically lead us.

    It has been pointed out that every gene used in the flagellum does something elsewhere in organisms. The cooption hypotheis says that all of the parts existed. Behe has suggested, well, so? Why would these parts assemble into something useless, waiting of the others to attach themselves to make a complete flagellum. I believe he is correct.

    In my opinion, for RM+NS to be true, a minimum of that minimal flagellum must have come to existance in a single mutational step.

    I therefore suggest that a search be made of all biological mechanisms that use any of the genes involved in the minimal flagellum. Surely this is a finite search. What must be found to prove that RM+NS is the plausible cause of the flagellum is for two of these mechanisms, not three, to contain all of the genes necessary to create a minimal working flagellum (They may contain more genes). If then genetic engineers can splice the two mechanisms together to create a flagellum, I believe that an adequate proof would be provided that RM+NS is a plausible cause for the flagellum.

    As this model limits itself to mechanisms which use the genes which are involved in the minimal flagellum, I believe that the search is finite. Once that search is complete, if no pair of mechanisms containing the genes necessary to create a minimal flagellum can be found, in my view RM+NS would be reasonably falsified as the probable cause of the flagellum.

    The only way that science could reasonably reconcile this discovery with RM+NS would be for some smarty to come up with two plausible mechanisms which have reasonable functional purpose, which do not exist and which, when combinded, produce a complete working flagellum.

    However, once the first level search is complete, I believe it is time for RM+NS to carry the burden of proof, to be considered incapable until proven capable.

Leave a Reply