Re: [asa] RE: (irreducible complexity and evolution) design and the nature of science (was: Re: [asa] Re: Gingerich on TE and ID)

From: Cameron Wybrow <>
Date: Tue Jun 16 2009 - 09:16:34 EDT


We are still not connecting. You write as if I offered this sentence --

"To disprove the claim that the flagellum is irreducibly complex, one would need to rip out a part of the flagellum and have it still function *as a flagellum* (not as a Type III secretory system)."

as the basis of an argument that, if the flagellum evolved via Darwinian means, it had to have evolved from a series of less effective flagella, and that as this is impossible, the flagellum could not have evolved by Darwinian means. But this was not what I had in mind when I wrote the sentence. It was intended only to show that your previous remarks (about the possibility of co-opting existing proteins) had not disproved the irreducible complexity of the flagellum, and in fact were based on a misunderstanding of Behe's use of the term "irreducible complexity". That is, *even if it is true* that the flagellum evolved from some earlier system (which could have been the Type III secretory system, or something else) by making use of ready-made sets of proteins, it does not follow that the flagellum, as we have it now, is not irreducibly complex.

As I said to Bernie, the question "Is this system irreducibly complex?" is distinct from the question "Could this irreducibly complex system have evolved via Darwinian means?" The notion of "irreducible complexity" is not by definition connected with the notion of "unevolvability" (in the way that the notion "having no legs" is by definition connected with the notion "unable to walk"). IC may have *implications* for evolutionary speculations of certain kinds -- Darwinian speculations -- but those implications aren't part of the definition. "Irreducible complexity" is an "engineering" description of the functioning of a system, not a statement about lack of evolvability. The "lack of evolvability" of irreducibly complex biological systems is something that requires further argument, beyond the bare statement of the definition of irreducible complexity.

Behe defines the term as follows:

"By *irreducibly complex* I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning." (DBB, p. 29)

That's his *definition*. What the definition means, in practical terms, is this: if you take the drive shaft or axle or battery out of your automobile, it won't run. And what he means by calling the bacterial flagellum an irreducibly complex mechanism is that if you remove any of the working parts, it no longer functions as a flagellum.
The *application* of this definition to Darwinian evolution is that you can't work up stepwise from earlier flagella to the current working flagellum, because if the current working flagellum is *truly* irreducibly complex (i.e., if Behe's engineering description of it is correct), then earlier flagella would by definition be missing parts without which the whole system could not function as a flagellum. Thus, he says:

"An irreducibly complex system cannot be produced directly (that is, by continuously improving the initial function, *which continues to work by the same mechanism* [emphasis added]) by slight, successive modifications of a precursor system, because any precursor to an irreducibly complex system that is missing a part is by definition non-functional." (p. 29)

Thus, the flagellum could not have evolved via Darwinian means from earlier flagella which were missing parts. That would be as logical as suggesting that our current automobiles evolved gradually from earlier mass-produced automobiles which were missing axles, batteries, or drive shafts. Such automobiles would never have been bought by the public, because they would not have worked, and therefore the companies which produced them would have gone out of business (natural selection), and then the current automobile would not have come into existence. I think you agree with Behe about this.

What Behe's definition does *not* exclude is the possibility that you and Ken Miller have mentioned, i.e., that a flagellum could have evolved from a system which originally served a *different* function, and served it *well*. For example, the Type III secretory system contains a major subunit of the flagellum, and that subunit serves a *different* function quite well. (I realize, by the way, that Ken Miller was not insisting upon the TTSS as an actual precursor, but was only offering it as an example of a working subset of flagellar proteins. Nonetheless, it could have served as an ancestral mechanism.)

However, the transition from the TTSS (or any other ancestral mechanism you choose to imagine) to the bacterial flagellum would have to be explained. Did the first system change in one gigantic shift into the second? That goes against Darwin's stipulation of gradualism; Darwin rejected such mega-adjustments in one generation as the equivalent of miracles and as something to be rejected by science. By steps, then? All right; but what steps? Did the TTSS (for example) keep working as a TTSS, while slowly accumulating parts it didn't need, to be later integrated into the flagellum when a final crucial part came along? That's logically possible; but in biological systems, things tend to be so tightly integrated that anything not necessary to the function will probably get in the way; so such extra, unused parts are very likely to gum up the TTSS mechanism in some way, or perhaps gum up some other function of the cell. Also, if such parts are unused for thousands of generations, they may themselves mutate away before they can be integrated into the flagellum. So what then? The most likely explanation is that by some tiny change the TTSS became some other functional system, neither TTSS nor flagellum, and then that system became still some other functional system, and so on, all the way to the flagellum. How many steps would that take? How can we tell, when we don't know what any of the hypothetical steps were or what genetic changes were involved in each? How can we tell whether they were genetically possible? How can we tell whether they were compatible with natural selection? How can we tell anything, when we are dealing with an unknown number of hypothetical transitional forms, each one of unknown genotype and phenotype?

This does not prove that there is no Darwinian pathway from some earlier system (TTSS or other) to the flagellum. It proves that such pathways are (a) extraordinarily hard to specify; and (b) entirely lacking in evidence. So why should one believe that any such pathway existed, unless one is dogmatically committed to the view that Darwinian mechanisms must be adequate?

To come back to the original point I was making to Bernie: Behe's *definition* of irreducible complexity does not, by itself, exclude the possibility that a precursor mechanism, using some of the same proteins, *which originally served a different function*, evolved into a new functional mechanism. But there are other, very strong reasons (given above) for thinking that such an explanation, while perhaps possible in the case of some biological systems, is very unlikely to be able to account for most macroevolutionary change. At least, it is very unlikely if the mechanism of evolution is the chancy one proposed by Darwin and championed by Dawkins and Miller. However, if something *else* is going on in the genome, i.e., if it is not just a string of nucleotides which randomly alter without rhyme or reason, but is a sort of master program for evolutionary change (as per fine-tuning, front-loading models), in which the overall pattern of mutations is not random but tends in certain directions, then such transformations may be quite possible and entirely explicable in naturalistic terms. Hence, the speculations of the anti-Darwinian Michael Denton, speculations which Behe has not ruled out. But again, such speculations are quite outside of conventional Darwinian thinking about isolated mutations that just *happen* to be useful for the next stage of evolution. They are speculations that the evolutionary process is a designed process. They are within the general framework of ID theory.

Once again for Bernie: irreproducible complexity is not incompatible with "evolution", but it poses a very great problem for "Darwinian evolution", i.e., the form of evolution that both Dawkins and Miller (and apparently a number of other TEs) believe in. Neither Dawkins nor Miller have come up with anywhere near a detailed Darwinian explanation for how an irreducibly complex system might be formed. What they offer instead is a superficially plausible notion of the co-option of systems previously used for other purposes. This process of evolutionary co-option is almost always presented in general conceptual terms, and is rarely accompanied by any of the nitty-gritty details. And that is characteristic of Darwinian evolutionary "science" generally -- it is full of broad general conceptions (mutation, drift, selection, co-option, etc.) which are rarely accompanied by nitty-gritty details. As long as Darwinians maintain this habit of speaking in grand generalities and avoiding all local mechanical specifications, I have no strong reason for believing that the evolution of any irreducibly complex system will ever be explained by Darwinian means.

  ----- Original Message -----
  From: Iain Strachan
  To: Cameron Wybrow
  Cc: asa
  Sent: Tuesday, June 16, 2009 2:40 AM
  Subject: Re: [asa] RE: (irreducible complexity and evolution) design and the nature of science (was: Re: [asa] Re: Gingerich on TE and ID)


  I don't have time right now to read Behe's rejoinders - I will attempt to do so. But I was responding to the argument as you stated it, on the fact that the existence of one intermediate doesn't explain the evolution of the flagellum - with which I agreed.

  If you wish to define irreducible complexity in this way:

>To disprove the claim that the flagellum is irreducibly complex, one would need to rip out a part of the flagellum and have it still function *as a flagellum* (not as a Type III secretory system).

  then you can't use it as an argument against evolution. It is only a valid argument under the assumption that the flagellum we see today must evolved in a stepwise fashion from a series of less effective flagella. If you insist on that then I agree that the irreducible complexity argument defies evolution. But the fact is that evolution doesn't work like that; proteins get re-used in other objects - it is not necessary to have them all present by a freak accident to get a functioning flagellum - they were all present and used in other functions. There is a New Scientist article on the flagellum not being irreducibly complex that states that of the 40 or so proteins, there are only two that are common to all flagella. Here's the quote:


  The best studied flagellum, of the E. coli bacterium, contains around 40 different kinds of proteins. Only 23 of these proteins, however, are common to all the other bacterial flagella studied so far. Either a "designer" created thousands of variants on the flagellum or, contrary to creationist claims, it is possible to make considerable changes to the machinery without mucking it up.

  What's more, of these 23 proteins, it turns out that just two are unique to flagella. The others all closely resemble proteins that carry out other functions in the cell. This means that the vast majority of the components needed to make a flagellum might already have been present in bacteria before this structure appeared.


  Emphasis mine. The irreducible complexity argument only has any force if one requires them all to appear at once by multple simultaneous mutations. But if the building blocks were already there it doesn't seem to me to have any force at all.

  I do not consider I have made an "honest error" and I also feel that your attribution of malice to Miller is not the sort of thing one Christian should be accusing another one of.

  I speak as one who was for a long while absolutely convinced of Intelligent Design, and who was convinced by reading Darwin's Black Box. If you read my posts to this listserve from around 2001 (they're not hard to find) you'll see someone who was anti-evolution. I have changed my mind since then.

  The reason I was anti-evolution was because I could see the irreducible complexity issue happening in the computer science field of genetic algorithms. Exactly that problem crops up if you try and design an evolutionary algorithm to solve an engineering problem. I reasoned that this meant that evolution itself could not work for the same reason. The error I was making was to assume that evolution worked in the same way as the genetic algorithm - the search for a specific solution. But it does not - there is no specific goal and components can get re-used in different ways.


  On Tue, Jun 16, 2009 at 12:40 AM, Cameron Wybrow <> wrote:


    You are quite wrong. You have not understood Behe's definition at all. You have made the same mistake that Ken Miller has made. The fact that some parts of an irreducibly complex machine can be used in another machine to serve a useful function does not prove that the first machine is not irreducibly complex. The fact that the wheel and axle of an automobile can be taken out and used for a go-cart does not prove that the wheel and axle is not a necessary functioning part of the automobile. The automobile (I mean the main parts thereof, not the radio or the leather seats) remains an irreducibly complex machine.

    To disprove the claim that the flagellum is irreducibly complex, one would need to rip out a part of the flagellum and have it still function *as a flagellum* (not as a Type III secretory system).

    Miller misreads Behe, I believe maliciously. In your case, I will assume it is an honest error. Please read Behe's rejoinders to Miller for further clarification, from the horse's mouth, if you don't trust my answer. They're easy to find.


      ----- Original Message -----
      From: Iain Strachan
      To: Cameron Wybrow
      Cc: asa
      Sent: Monday, June 15, 2009 5:29 PM
      Subject: Re: [asa] RE: (irreducible complexity and evolution) design and the nature of science (was: Re: [asa] Re: Gingerich on TE and ID)


      Regarding the Type III Secretory System.

      I don't think the point of Miller's example is to "solve the problem that Behe is raising" as you put it. It doesn't show us how the flagellum evolved & indeed we don't know this. The point is whether or not the flagellum is irreducibly complex.

      Miller has demonstrated the existence of smaller parts of it in another functioning organelle, and thus shows that it is _not_ irreducibly complex.

      I think that's QED. The whole case rests on the idea that the flagellum is irreducibly complex. It isn't.

      The key point is that the smaller part of it is used for a different function, but one that also has benefit.


      On Mon, Jun 15, 2009 at 9:33 PM, Cameron Wybrow <> wrote:


        Bill has partly answered your posts already, and my understanding is about the same as his, but let me try to address your objections in my own way.

        Two things must be kept distinct:

        1. The nature of an irreducibly complex system
        2. The question whether Darwinian evolutionary processes can produce irreducibly complex systems

        On point 1:

        "Irreducible complexity" is a non-historical, functional description of a system. It indicates that all the parts are necessary for the whole to function. In an irreducibly complex system, all the parts interact in such a way that the removal of one of them would cause the system to cease functioning. This is basically an engineering concept, having nothing intrinsically to do with evolution or even with biology.

        On point 2:

        We know from our experience as human beings that irreducibly complex systems can be built by intelligent agents, e.g., ourselves. We do not have experience of inanimate matter building irreducibly complex systems. (Unless the inanimate matter is following a template, e.g., as in embryonic development; but I am speaking of building them up from scratch, or from earlier systems which are quite different.) So the question is, can natural processes, unguided by any intelligence, produce irreducibly complex systems, or transform one irreducibly complex system into an entirely different one?

        Darwinian evolution says that they can. Behe's argument is that there are formidable barriers in biological systems to the production of irreducibly complex systems by means of accidental mutations, even when those mutations are culled by natural selection, because natural selection can't select for a system which does not function. So how do you get enough mutations together to make up the new system that you want, so that "natural selection" can select it?

        Behe grants that in theory there might be intermediate stages between two irreducibly complex systems, each of which has survival value. But he doesn't find much evidence in the Darwinian literature of these stages.

        Regarding Ken Miller, Miller and Behe have been at this for years. They have published several articles each on the irreducible complexity of the flagellum. I think it is important that if you read Ken Miller's criticism of Behe's argument, you read every one of Behe's rejoinders.

        First, Miller sometimes misrepresents Point 1 -- what Behe means by an irreducibly complex system. Whether this is out of lack of understanding or out of malice, I don't know. Anyhow, Behe clarifies what he means in his rejoinders. You can read them.

        Second, Miller has suggested as a possible intermediate the Type III secretory system. Let's say we grant this. It still doesn't do nearly enough to solve the problem that Behe is raising. There would have to be a series of useful intermediates, not just one, between your basic bacterium and your bacterium with a flagellum. So you still need some notion of what these stages would have been, how they would have given the creature survival advantage, how they would be generated genetically, etc. 99% of the story is missing. And in many cases in the animal kingdom, 100% of the story is missing. What are the intermediate stages? What survival value did they have? An eagle with 5% sharper eyesight has a definite survival advantage. But what about a bat with half a wing, enough to impair the normal functioning of the arm, but not good enough to fly, or even glide in a stable manner? What survival advantage would that intermediate form have? Or a malfunctioning sonar, one which sometimes caught mosquitoes very well, and other times let the bat crash into the cave wall? What survival advantage would that intermediate form have? If evolution had to work up through such imperfect partial stages, it would certainly never get to the bat. And Darwinists have yet to come up with even a single plausible sequence of intermediate forms, anywhere in the animal kingdom, that could produce complex organic machinery such as the cardiovascular system.

        We can trace the "evolution" of the motor car. We can see "intermediate stages" between the Model T and modern cars. But all of those intermediate stages were designed, irreducibly complex systems. They did not mutate into the next phase by accident. The question is, can nature do the equivalent, and manage, not just frequently but millions of times, to find its way, unguided, to useful intermediate forms? The Type III secretory system, one possible intermediate in a single type of unicellular creature, is very thin evidence upon which to rest a mechanism which is supposed to explain the entire macroevolutionary process.

        To bring this back to your original objection: Behe allows that "evolution" (not of the Darwinian kind) might have produced irreducibly complex systems. But if so, it would be "front-loaded" evolution of some kind, in which the parts were, so to speak, self-assembling, almost as if they had an inbuilt intelligence which told them when to appear and where to take their place in the new arrangement. Needless to say, such a front-loaded evolution would require a mind with an ability to handle complex interactions far beyond that of the greatest supercomputer, a superintelligence which could only be divine. So if you go for front-loading, you have to believe in God. Maybe not the Christian God, but some kind of God. Thus, Denton, who proposes such a system, speaks of God. Behe stops short of endorsing Denton's view as the correct one, but allows it as a possible means of producing irreducible complexity through an evolutionary process.

        Once again, unless you rigorously separate "evolution" from "Darwinian evolution", as I have been begging everyone here to do, you will never understand the arguments or motivations of ID proponents. The intelligent design of irreducibly complex systems is logically compatible with evolution, but not of course with Darwinian evolution.


        ----- Original Message ----- From: "Dehler, Bernie" <>

        Cc: <>
        Sent: Monday, June 15, 2009 2:18 PM
        Subject: RE: [asa] RE: (irreducible complexity and evolution) design and the nature of science (was: Re: [asa] Re: Gingerich on TE and ID)

          Bill said:
          "The issue is the number of elements that must be simultaneously
          available. If all the parts of a mouse trap are present simultaneously
          it might even be possible for a car accident to put it together, but
          highly unlikely."

          As I understand irreducible complexity (IC) , it is argued that the parts can't al be around at the same time because there is no use for them to be around. That is the whole point of the argument- since there is no use for them, they must have been created together all at once. Therefore- evolution, guided or unguided, is impossible. Evolution is all about building something more complex from more simpler building blocks. IC is all about this evolution being impossible, because there is no need for the lower building blocks to be around.


          -----Original Message-----
          From: Bill Powers []
          Sent: Monday, June 15, 2009 10:34 AM
          To: Dehler, Bernie
          Subject: Re: [asa] RE: (irreducible complexity and evolution) design and the nature of science (was: Re: [asa] Re: Gingerich on TE and ID)


          I'll let Cameron reply for himself.

          But clearly your understanding of irreducible complexity cannot be
          correct. After all, we do construct mouse traps and cars all the time.

          The issue is the number of elements that must be simultaneously
          available. If all the parts of a mouse trap are present simultaneously
          it might even be possible for a car accident to put it together, but
          highly unlikely.

          It is a matter of probabalistic resources. In order to put togehter a
          car, say on an assembly line, it is required that
          1) the available parts be simultaneously, or more carefully that they
          would be available on the time scale much shorter than the dissipative
          time scale (i.e., some measure of the effects of entropy, friction, or
          chaotic mixing).

          2) That all the pieces would be put together on a time scale short
          compared to the disordering time scale (e.g., the time for the parts to
          rust, etc.).

          Irreducible complexity, as I understand it, is a statement that the
          joint probability of both the parts being nearly simultaneously
          available, say over time DeltaT, and the probability of the parts being
          put together over a time DeltaT is vanishingly small for a non-lawful,
          non-intelligent process.

          It assumes that there is no necessity in the parts coming together
          (e.g., the parts of a mouse trap have not "lawful" reason to come
          together, although once together they remain together, implying, it
          seems short range forces). The parts must have a lifetime that is long
          compared to the time of preparation, otherwise what is put together will
          not be made of the parts, although it might be possible that even a
          rusty car could become a whole, or an "inferior" mousetrap.


          On Mon, 15 Jun
          2009, Dehler, Bernie wrote:

            Cameron quoted:
            " If the parts of the universe are so fine-tuned that they cannot help but fall together into certain biologically functional arrangements, then irreducibly complex structures will of course form through a front-loaded evolutionary process."

            I don't understand. I understand "irreducible complexity" to mean that it is impossible for nature to sequentially build something complex because all the parts would need to appear at once (like the mousetrap needing all of it's basic parts to be present at the same time). This is against evolution- guided or unguided.


            -----Original Message-----
            From: [] On Behalf Of Cameron Wybrow
            Sent: Monday, June 15, 2009 7:35 AM
            Subject: Re: design and the nature of science (was: Re: [asa] Re: Gingerich on TE and ID)

            Actually, Bernie, that's not the case. Here are just three paragraphs for

            If the parts of the universe are so fine-tuned that they cannot help but
            fall together into certain biologically functional arrangements, then
            irreducibly complex structures will of course form through a front-loaded
            evolutionary process. The evolutionary process would then be analogous to
            the embryological process, in that the outcome would be planned from the
            beginning, with the changes all ready in advance. But this would be a
            planned evolutionary process, unlike Darwin's evolutionary process, the
            essential nature of which is to be unplanned. Irreducible complexity does
            not clash with "evolution"; it clashes with "unplanned evolution". Behe
            assumed that any careful reader of *Darwin's Black Box* would understand
            that he was talking about "unplanned evolution".

            Of course, there is as yet no evidence that fine-tuning can accomplish such
            an evolutionary feat. Fine-tuning at the evolutionary level essentially
            requires that the earliest DNA has the plan for the whole march of life
            implicit in it, and this claim is still highly speculative, and will remain
            so until we understand what all the apparently unused DNA is for. But this
            view is no more speculative than Darwinism itself. Each view requires one
            seemingly incredible claim. Darwinism requires us to believe that time
            after time, over the course of life, accident produces new functional
            structures, with the whole process conveniently tending upwards toward man;
            front-loading or fine-tuning requires us to believe that level after level
            of biological complexity is packed into the fundamental properties of the
            chemical elements, and this gives to biochemical necessity the awesome power
            that Darwinism gives to chance. On the face of it, it seems incredible that
            either chance or necessity could have such powers.

            The smart money, I would say, is on the view that the evolutionary process
            is in fact intelligently guided. It's a view consistent with our knowledge
            of integrated complex systems, and of the limited power of Darwinian
            mechanisms. It's a view in tune with common sense. It's the view that
            seems to be held by Ted Davis and George Murphy and Robert Russell, and
            also, I am told, by Owen Gingerich and other TEs. It is also a view
            compatible with the arguments of Behe and of many other ID supporters. But
            it is of course a view which, according to many here, is unscientific, or
            non-scientific. I never said that intelligently guided evolution was
            "scientific" in the sense of a formal scientific hypothesis, but I certainly
            believe that it is compatible with the best scientific data -- more
            compatible, in fact, than pure Darwinism is. Once again, I think that TE,
            when it is not afraid to say directly and without ambiguity that God
            controlled the evolutionary process to produce certain specific results, is
            in tune with some versions of ID.


            ----- Original Message -----
            From: "Dehler, Bernie" <>
            To: <>
            Sent: Thursday, June 11, 2009 3:19 PM
            Subject: RE: design and the nature of science (was: Re: [asa] Re: Gingerich
            on TE and ID)

              I think one TE could say "yes" (fully gifted creation- Howard Van Till),
              and another might say "No" because God needs to guide it somewhat or at
              certain times.

              Cameron said:
              ID says, "NO -- unless the properties of the atoms and the laws of nature
              are fine-tuned, which implies design."

              If ID says it could happen, then that blows apart Behe's mousetrap example
              (irreducible complexity), which is a pillar for ID I think ("Icons of ID"
              if you will).

            To unsubscribe, send a message to with
            "unsubscribe asa" (no quotes) as the body of the message.

            To unsubscribe, send a message to with
            "unsubscribe asa" (no quotes) as the body of the message.

          To unsubscribe, send a message to with
          "unsubscribe asa" (no quotes) as the body of the message.

        To unsubscribe, send a message to with
        "unsubscribe asa" (no quotes) as the body of the message.

      Non timeo sed caveo
      (")_(") This is a bunny copy him into your signature so he can gain world domination

  Non timeo sed caveo
  (")_(") This is a bunny copy him into your signature so he can gain world domination

To unsubscribe, send a message to with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Tue Jun 16 09:17:20 2009

This archive was generated by hypermail 2.1.8 : Tue Jun 16 2009 - 09:17:21 EDT