Posted by: rmbrowning | June 12, 2009

Telling tails about bacterial flagella

Note: This post may be subject to some revision. I’ve hopefully managed to accurately summarise some scientific findings outside my area of proficiency.  Also, there is a lot that could be said here but I’m be trying my best not to bog this down too much. Dialogue and suggestions are most welcome.


The bacterial flagellum (from latin “whip”) is one of the most complicated examples of biological “machinery” we know of today. The projecting filament of the flagellum is capable of rotating at thousands of revolutions a minute to propel bacteria through watery environments, in a similar way to an outboard motor. Explaining how such a system works and how it evolved has been a subject of interest for modern biologists.


In recent years, the interest in this curious piece of micro-machinery has increased dramatically. This has been as result of the intelligent design (ID) movement using it as a challenge to evolution and the flogging the subject took at the Dover trial a few years ago. Cdesign propensists, claim that the bacterial flagellum is evidence for ID as it exhibits an attribute called “irreducible complexity” (IC), which was coined by Michael Behe.


What he means by this is:


“An irreducibly complex system cannot be produced directly by numerous, successive, slight modifications of a precursor system, because any precursor to an irreducibly complex system that is missing a part is by definition nonfunctional.”

Since natural selection can only choose systems that are already working, then if a biological system cannot be produced gradually it would have to arise as an integrated unit, in one fell swoop, for natural selection to have anything to act on.”


Essentially, an irreducibly complex system could be formed by the gradual processes of evolution as the non-functioning intermediates would not be preserved by natural selection. This harkens back to the old creationist arguments of “what good is half an eye” and William Paley’s argument for design (Perhaps these arguments and IC share a common ancestor?).


Despite the over-whelming rejection of ID by the scientific community and the gaping wound left by the rulings at Dover, ID proponents insist they have a case for design in irreducible complexity.  Do they?


I don’t think so when I consider:

  1. the bacterial flagellum isn’t irreducible

  2. homology within the proteins involved

  3. the many paths to bacterial mobility

  4. complexity design

An article by Nick Matze mentions that some bacterial flagella can function without some of the proteins that make up the flagellum. According to the definition we should not be able to reduce such a system to a functioning simpler form. It should be by definition non-functional.


Another reason not to accept the notion of irreducible complexity is the existence of homologies. One popular example is that of the type 3 secretory system (T3SS) which pathogenic bacteria use to inject cells with toxins. Several of the T3SS proteins are found in the bacteria flagellum. It’s currently debated which came first or whether they evolved from a common ancestor, regardless, this example shows how a partial flagellum could have a use. If it has a use it can be preserved by natural selection and exaptation/co-option to act on it, which leaves no reason to claim evolution is incapable of producing it.


In reading some of the literature surrounding bacterial flagella I came across something I had previously overlooked, which is possibly my new favourite reason to reject ID’s notions. The fact is that there is no such thing as the bacterial flagellum. Instead, we find huge variation in how flagella function and are assembled. Of the 40-50 proteins used in making a flagellum only 20 are universal. I think it’s unnecessary to detail all the possibilities, but this raises some rather awkward questions for the ID camp such as:


Why did the intelligent designer reinvent the design so many times?


Did the intelligent designer create all of these varying different flagella in thousand of special creations?


Is the intelligent designer trying to trick us by making it look like these different systems evolved from a common ancestor?

The final question for me is how on earth would we recognise design? Complexity is not necessarily a good indication that something is designed. What about simplistic designs? Paperclips and bricks aren’t complicated but we know that they are designed (albeit by human designers which is another limitation for intelligent design). What we’re actually looking for in design is artificiality, which is difficult to find in a biological system.


Bacterial flagella are complicated and their evolutionary history is not well understood.  However, I think that it is much better to leave science to explain, rather than invoking the arguments from psuedo-science which halt scientific enquiry.  Pointing at gaps in scientific knowledge without providing a scientific and testable alternative is not good science.  Furthermore, intelligent design implicitly limits God’s intelligence and power (most ID proponents think of the Judeo-Christian God).  A God who has to repeatedly intervene to maintain his creation is less intelligent than one who could had a process in mind which could produce such complex micro-machines.


Relevant articles:


Dolittle, W.F. & Zhaxybayeva, O., (2007). Evolution: Reducible Complexity — The Case for Bacterial Flagella , Current Biology, 17(13) , 510-512

Pallen, M.J., & Matzke, N.J., (2006). From Origin of Species to the origin of bacterial flagella, Nature Reviews Microbiology, 4, 784-790

Wong, T., Amidi, A., Dodds, A., Siddiqi, S., Wang, J., Yep, T., Tamang, D.G., & Saier, M.H., (2007). The Evolution of Bacterial Flagellum, Microbe, 2(7), 335-340

Advertisements

Responses

  1. “Why did the intelligent designer reinvent the design so many times?

    Did the intelligent designer create all of these varying different flagella in thousand of special creations?”

    So, basically, your argument here is “I don’t think a designer would have done it this way.”

    I do not see a problem with a common designer using somewhat different designes. Designers do this all the time.

    Just because there are there are flagella with certain differences , that doesn’t prove they weren’t designed. — Henry Ford used different designes on automibiles when he worked on newer models, but there is no dispute that they were all designed. — Using somewhat different designes therefore does not disprove a common designer.

  2. So, basically, your argument here is “I don’t think a designer would have done it this way.”

    Partially. The real crux of it is though, why make it look like all these flagella evolved from a common ancestor? Accepting thousands of special creations over evolution is straining Occam’s razor a bit.

    I do not see a problem with a common designer using somewhat different designes. Designers do this all the time.

    Human designers do this all the time. Do we even know how this intelligent designer is supposed to work?

    Just because there are there are flagella with certain differences , that doesn’t prove they weren’t designed. — Henry Ford used different designes on automibiles when he worked on newer models, but there is no dispute that they were all designed. — Using somewhat different designes therefore does not disprove a common designer.

    Your analogy is a bit flawed.

    Firstly, Henry Ford was a human designer. We can recognise human work and we have a history confirming his work.

    Secondly, cars are artificial non-reproducing mechanical constructions so reaching the design conclusion is easy.

    What about biological systems that reproduce, undergo mutation, gene duplication, co-option etc? We don’t have any scientific evidence or mechanisms for this design. Instead we have multitude of different “designs” and various homologies. Why would we assume it was designed when it has the appearance of evolution?

    • “Partially. The real crux of it is though, why make it look like all these flagella evolved from a common ancestor? Accepting thousands of special creations over evolution is straining Occam’s razor a bit.”

      It is possible (and perfectly acceptable in an ID/Creation view) that some of the variation and length of different flagella have resulted from mutation, but that has nothing to do with Irreducible complexity because they do not effect the “well matched and fitted” components which actually make up the irreducible system. — The fact is that they still follow the same basic design which is perfectly acceptable in an ID/Creation view which fully accepts that certain change “within a kind” does occur.

      So, nobody is claiming that there has to have been thousands of special creations. This is perfectly explainable as “variety in a kind” which is like variety in the dog kind. (Chiuahua vs. Poodle which both had a common ancestor)

      “Firstly, Henry Ford was a human designer. We can recognise human work and we have a history confirming his work.”

      True, but there is no reason to assume that a non-human designer wouldn’t have used the same kind of principles by design.

      “Secondly, cars are artificial non-reproducing mechanical constructions so reaching the design conclusion is easy.”

      I really don’t see how this invalidates the argument. Made-made designes are common analogies that ID proponents use, for example, Paley’s watch and Behe’s mousetrap. I don’t see how being non-reproductive changes the analysis.

      • krissmith777,

        Chiuahua vs. Poodle which both had a common ancestor

        What evidence do you have that Chiuahuas and Poodles had a common ancestor?

  3. ryan,

    very nice overview of the topic. you captured its essence well.

    krissmith,

    i have a couple of issues with your arguments. to start at the end, Behe’s mousetrap IS reducible. for an exploration of this (which is also quite fun), check out this website: http://udel.edu/~mcdonald/mousetrap.html
    A common fault of ID folks is their use of faulty analogies.

    secondly, not only do flagella have plausible evolutionary pathways, they also are reducible in their current state. The journal article below discusses flagellin, the protein which forms the filament in many types of flagella and which is present in nearly all flagellated bacteria. Kuwajima was able to remove one third of the 497 constituent amino acids of flagellin without seriously harming the function. So tell me exactly how that is irreducibility? Just because something is well matched and fitted does not mean something else could not provide exactly the same capability. (Kuwajima, G. 1988. Construction of a minimum-size functional flagellin of Escherichia coli. Journal of Bacteriology 170: 3305-3309.)

    Behe’s arguments are often based on the fact that he simply could not think of or find examples of reducibility (though he could have found many examples if he had looked). They are not often based upon experimentation which SHOWS evidence of irreducibility. All of the actual evidence is to the contrary.

    • jonathanwaldroup,

      “i have a couple of issues with your arguments. to start at the end, Behe’s mousetrap IS reducible. for an exploration of this (which is also quite fun), check out this website: http://udel.edu/~mcdonald/mousetrap.html
      A common fault of ID folks is their use of faulty analogies.”

      I am fully aware of John McDonald’s arguments. And I am also fully aware that he completely misses the point.

      All he shows is that is is possible to construct a moustrap with less parts. — Michael Behe never claims that it cannnot be. His argument is that the 5 part mousetrap cannot work without all the pieces, not that it is impossible to make one with less pieces.

      And if I may say so, the traps that function with less pieces are designed to work with less.

      “The journal article below discusses flagellin, the protein which forms the filament in many types of flagella and which is present in nearly all flagellated bacteria. Kuwajima was able to remove one third of the 497 constituent amino acids of flagellin without seriously harming the function.”

      Common argument which also misses the point. Michael Behe never says every protein absolutely has to be in place.

      On page 39 when Behe talks about pieces having to be in place and in the right order, he specifically mentions that he is talking about the well-matched, interacting ones that contribute to the basic function. — Behe never says every single protein is part of that system.

      Irreducible complexity is based on a minimum of pieces in the right order, not on a maximum.

      I discuss misconceptions of ID here: http://explanationblog.wordpress.com/2009/06/09/evolutionists-ignorance-about-intelligent-design/

      • krissmith777

        “His argument is that the 5 part mousetrap cannot work without all the pieces, not that it is impossible to make one with less pieces.”

        Many critics have constructed functioning 3 and 4 part mouse traps. The point is that if simpler versions of the mechanical devices could work, then simpler versions of biochemical systems could work as well. Which means that these complex biochemical machines could have had functional precursors.

        Also, Ken Miller uses part of mouse trap as a rather crude tie clip. No, it’s not working as a mouse trap but it’s fully functional in another context. If it has a function natural selection can preserve it and co-option, gene duplication etc can put parts together to make something else.

        On page 39 when Behe talks about pieces having to be in place and in the right order, he specifically mentions that he is talking about the well-matched, interacting ones that contribute to the basic function. — Behe never says every single protein is part of that system.

        Where does “irreducibility” start and finish then? Many of the parts contributing to the basic function can be removed or are absent from certain bacteria, but the resulting organelle still functions. Is that not an example of reduced complexity or did the goal posts move? In most other cases, around 3 other proteins can be used instead to create one of these well matching parts. Again, why create all of these homologies?

      • Kris,

        Flagellin IS absolutely necessary to the functioning of the flagellum. Of all the components, this is the protein that is absolutely vital. It is the largest constituent piece – surely you include the largest, most important proteins in your definition of irreducibility. Not even Behe would doubt the importance of flagellin.

        The problem with your argument is that irreducibility is not well-defined, at least, not the way you are using it. I honestly think Behe felt that every protein in the flagellum was “well-matched” and “interacting.” He certainly implies that every piece of the flagellum is needed. It was not until after he proposed his theory that our understanding of the flagellum allowed us to realize certain parts were unnecessary. But using the argument you use, you could argue for irreducibility on anything, arguing that as we discover more and more things that can be omitted, that really the irreducible system was the newly discovered system without all of its extraneous pieces. We could do that all the way down to a two-piece system, at which point it would be extremely easy to explain the appearance of a system on an evolutionary basis and the system would no longer be irreducible.

        You are calling the flagellum irreducible because you want it to be, but you are incapable of actually saying what the irreducible parts are! If you can show me what the actual irreducible system is, then we will work from there and see if that system is actually irreducible. But neither you nor Behe does this, and as I said, Behe implies that the entire flagellum is irreducible, but that is clearly false now. So irreducibility is not testable. It assumes there is something irreducible, but it waits for science to eventually find that smallest component that really is irreducible, and that is always getting smaller and smaller.

  4. “Many critics have constructed functioning 3 and 4 part mouse traps. The point is that if simpler versions of the mechanical devices could work, then simpler versions of biochemical systems could work as well. Which means that these complex biochemical machines could have had functional precursors. “

    That is true, but nobody disputed that simpler versions could exist.

    “Also, Ken Miller uses part of mouse trap as a rather crude tie clip. No, it’s not working as a mouse trap but it’s fully functional in another context. “

    What he did was say that an incomplete mousetrap could be used as a paper-weight. Again, that is not disputed. Nobody claims that there are no other uses at all, least of all Michael Behe. What he means is the basic finction that it is used for, not every single lesser function.

    “Where does “irreducibility” start and finish then?”

    I have already mentioned that he specifically mentions the well-matched, interacting parts. —

    Think about it this way. I the case of a car, you can remove a bumber, a hood, the head-lights and still a car in which the basic funtion has not been impeded.

    That is because the “irreducibility” comes down to the car having a working engine, a steering wheel, breaks, and the four wheels. — As I said before, this irreducibility is based on a minimum of parts. The wheels and the engine are thewell-matched, interacting parts. The rest are definitely beneficial, but not essential for the basic function.

    Saying that “irreducibility” indicate a maximum amount of parts is like saying that a car cannot properly run without a bumber. But that is not the case.

    “Many of the parts contributing to the basic function can be removed or are absent from certain bacteria, but the resulting organelle still functions. “

    That is true, but they are not essential to the basic function. They have their benefits, but they aren’t necessary.

    It has been said, “If X is an irreducibly complex process, adding Y to improve it does not make the result not irreducibly complex. It just means that Y is not part of the basic function.”

    • Sorry, for the late reply I’ve been busy. I’m a bit tired but hopefully this is reasonably coherent.

      “That is true, but nobody disputed that simpler versions could exist.”

      Ok, so you agree we can have functional precursors. Why assume that numerous, successive, slight modifications of these precursors could not produce a more complex system?

      “What he did was say that an incomplete mousetrap could be used as a paper-weight. Again, that is not disputed. Nobody claims that there are no other uses at all, least of all Michael Behe. What he means is the basic finction that it is used for, not every single lesser function.

      Why assume then, that it could not evolve? If we have examples such as the T3ss which show how a partial flagellum could have a use and we know that co-option/gene duplication could produce new functions, why invoke a non-scientific explanation without a mechanism?

      ‘I have already mentioned that he specifically mentions the well-matched, interacting parts.”

      In the case of the bacterial flagellum, which parts are those?

      “That is because the “irreducibility” comes down to the car having a working engine, a steering wheel, breaks, and the four wheels. — As I said before, this irreducibility is based on a minimum of parts. The wheels and the engine are thewell-matched, interacting parts. The rest are definitely beneficial, but not essential for the basic function.”

      These artificial mechanical analogies aren’t really very good but I’ll play along for once. I’m not much of a mechanic but how about this for reducing a car:

      1. Knock out a couple of cylinders of the cars engine
      2. Remove the brakes entirely
      3. Strip the tyres off the wheels

      The car still goes but, it has less power, can’t really stop, is woefully inefficient and creates an incredible mess. A lot is left to be desired. However, it’s still mobile which confers an advantage over non-motility. If mutations, co-option and gene duplication are able to create more cylinders or rudimentary tires, perhaps after thousands of generations we’ll have something that resembles a modern car. Perhaps, with more of this in a few more generations we’ll have a rally car with four wheel drive, active yaw control, a turbo-charger and a carbon fibre spoiler. Such as a Mitsubishi Lancer Evolution VIII. (Sorry that was awful, I just like evo’s.)

      This example is better. (Promise.)

      Or how about this to show that just because something is essential in it’s present form, it doesn’t mean that it always was. Some cars have GPS and it’s not currently essential. You’d say that it’s not part of “the well matched parts”. Fine.

      What about if after years of car evolution we no longer need drivers. The cars run using satellites, and computers control steering etc. If people of the future looked at a car they’d say that GPS is absolutely essential as the car is non-functional without it.

      Complex biological systems are the same, co-option may take something used for a beneficial (albeit non-essential) function and make it into an essential component. Even if it’s essential now, doesn’t mean it always was.

  5. “Is that not an example of reduced complexity or did the goal posts move? “

    No, because not every protein is part of the actual complex system itself.

    “most other cases, around 3 other proteins can be used instead to create one of these well matching parts. “

    True, but irrelevant.

    Actually, it is freely admitted by ID proponents like Casey Luskin, Stephen Meyer, and many others.

    “Again, why create all of these homologies?”

    Why assume that homology is evidence of common descent? The argument of homology is unreliable. For example, one science article in “New Scientist” shows that, despite similar homology, horses are not as closely related to cows as they are to bats. (http://www.newscientist.com/article/dn9402-bats-and-horses-get-strangely-chummy.html)

    Another example is that of North American wolves and the tasmanian wolf. Despite similar homology, the two are not closely related. So, similar homology is not conclusive evidence of common ancestery.

    • Ryan, you may be interested to note that krissmith777 has come to the conclusion that the various scientific claims regarding evolution are, in fact, true. He’s got a new website that I stumbled across a couple of months ago. I only mention this because it often feels like we’re trying to nail jelly to the ceiling in attempting to engage with creationists and this discovery was encouraging. You may find it encouraging also. Well done.

      • Thanks very much for that Damian! That’s excellent news! I’ve been trying to do a few more posts and I had thought about discussing Genesis/my beef with YEC. This gives me a bit of motivation to get around to it sooner. Believe me when I say that you’re not the only one who is dismayed by Christian theists and their reluctance to honestly evaluate mainstream scientific findings. I promise we’re not all that bad.

        Blessings,

        Ryan

  6. I read your original post with care, and planned to make a few small points–but I see that better people than I have already addressed them.

    You don’t seem to understand “intelligent design” to mean the same thing that I do by the term. I don’t know you well enough to accuse you of fallacies like “setting up a straw man,” but it seems to me that you’re critiquing a caricature of ID. If you really believe you’ve been fair to ID, as you understand it, then I’d suggest you reread some of the comments above–they’re hitting on the points I would have made.

    In discussing ID, it may be useful to use the analogy of prime number cryptography. Some code systems use the prime factors of very large numbers to conceal information. As you may know, it is very hard to find all the prime factors of a very large number–you have to divide the number by every possible factor up to the square root of the number. When you have a number with a few hundred digits, that means you have to do as many long division operations as there are particles in the universe.

    If I showed you all the prime factors of a number with 300 digits, you could multiply that number by two and produce a new number and all its prime factors. Multiplying by two isn’t hard. In the same way, if I could produce all the proteins for a working flagellum, I’ll bet it wouldn’t be hard to add one more amino acid to make it work a little better. “Irreducible complexity” is to “actual current proteins” as “the prime factors of my original 300 digit number” is to “the prime factors of two times my original 300 digit number.”

    The ESSENCE of the irreducible complexity argument is that it’s REALLY hard to try all the combinations you need to get the first biomechanical machine to work. If you think the number of trial and error operations needed approach that of prime number cryptography, you’re likely to find ID arguments persuasive. If you think the number of trials is well below that, you’re likely to find it laughable.

    So… the human response to the ID argument turns out to depend on just how many trials it takes to get a flagellum working. That would seem to be a mathematical question that can be asked and answered.

    Has anybody computed the statistical possibility of randomly assembling amino acids into a working flagellum?

  7. Has anybody computed the statistical possibility of randomly assembling amino acids into a working flagellum?

    When living things evolve it isn’t a random process. Beneficial mutations are conserved and passed on to future generations. But you already know that eh?

  8. The ENTIRE argument about “irreducible complexity” is that there appear to be some systems that arguably have no “beneficial” effects until they are complete.

    The best example of this is the first molecule capable of copying itself despite modifications. John von Neumann called these ‘cellular automata.” Viewing these as a kind of “software,” he computed how long the shortest possible program would have to be that could copy itself after suffering some sort of “mutation.”

    I’ve read through the notes from von Neumann’s lectures on this topic, which he delivered at Harvard University in the 1950s. I couldn’t follow his math, I’m sorry to say, but he came up with 1400 bits as an answer. In theory, then a self-reproducing molecule is “irreducibly complex,” and has a minimum length of 1400 bits.

    • “The ENTIRE argument about “irreducible complexity” is that there appear to be some systems that arguably have no “beneficial” effects until they are complete.”

      Exactly, and a partial flagellum could easily have a beneficial function. Organelles such as the type 3 secretory system show how a partial flagellum could have a function for natural selection to preserve.

      • Ryan!
        just stopping by and saw this – good stuff here.

        In my talk (you might remember) at the Faithful Science thing, I also mentioned that the TTSS is an example of a useful precursor to a flagella. My remark was picked up by the illustrious (and colourful) Dr. Robert Mann, who (though he himself is harshly critical of I.D.) wanted to make known that the TTSS is from a much later organelle than that of the flagella, apparently, making a link well-possible from the complex flagella to the TTSS. It remains that something like the TTSS would have been the precursor to to the flagella, but just not that particular one.. That is all 😛 More posts please!

  9. I’d be interested to see the same calculations applied for turning a mustard plant into a broccoli plant or a wolf into a poodle. Or perhaps for a bacterium to evolve the ability to break down nylon or citrate.

  10. The calculations that I’m describing are only helpful when you’ve got a situation where there is NO evolutionary benefit to an intermediate stage. So the mustard->broccoli situation doesn’t fit. There are probably many useful steps between mustard and broccoli where selection pressure would conserve changes.

    rmbrowning, what I hear you to be saying is that some subcomponents of the flagellum could have had an evolutionary value on their own. That reduces the challenge of putting together the complete flagellum. It’s as if carburetors, fuel pumps, and pneumatic tires all had their own uses long before an automobile ever appeared. It would be simpler to arrange those preexisting pieces into a working car than to come up with all of them all at once.

    I was more intrigued by the blood clotting example of irreducible complexity. There are probably some good answers to this one, but the argument is that it takes 20 separate proteins too cause blood to clot. If any one of those proteins is missing, blood doesn’t clot–which leaves the creature without a clotting agent always at risk of bleeding to death. On the other hand, if all 20 of those proteins don’t work perfectly, the blood clots when it shouldn’t, killing the poor little mutant with a stroke. How do you get all twenty right the first time, if any error either way results in death?

  11. This highlights perfectly why ID is not a science. It’s essentially an argument from incredulity combined with a God-of-the-gaps argument. If science ever discovers every step of the evolution of the bacterial flagellum showing that every stage can be accounted for by natural processes you’ll just move on to another area that we don’t have a complete knowledge of.

    Can’t think of a way to get from mustard to broccoli? Goddidit. Found an answer? OK then, how about the bacterial flagellum? Type 3 secretory system you say? Very well then, what about blood clotting?

    And so it goes.

    It’s a disingenuous and skewed form of argument that is unwilling to apply the same set of requirements to it’s own claims. So, let’s say that everything was evolving nicely but along came a hurdle and God really wanted some bacteria to have outboard motors. Please, explain the mechanism for how God caused them to suddenly develop this. Bueller? Bueller?

  12. Damian,

    Thank you very much for the above comment, you’ve pre-empted a lot of what I wanted to say.

    Scott,

    Yes, I think you’ve got the gist of the points I was making about flagella. With all the homologies and potential for functional subsystems, I don’t see a good reason to posit that flagella are un-evolvable.

    It seems as though abandoned the flagella now and moved onto blood clotting? This is one of my issues with the “God of the gaps” in intelligent design. If we begin to understand one of these gaps God gets squeezed out, which is potentially very detrimental for peoples faith.

    I’m probably not the right person to ask about blood clotting. I did a pathology paper a few years ago and it’s a nightmare to understand by itself, let alone how it evolved. I’ve had a quick skim over Ken Miller’s stuff on the subject and there are certainly things to be said.

    A lot of the clotting factors are amplifying the cascade. You’ll still clot without them but it takes longer for it to form. Obviously even this lesser clotting is advantageous over not clotting at all so it could be preserved by natural selection. Some other mammals, such as dolphins, lack at least one of our twenty clotting factors and get along fine.

    Obviously, we didn’t just evolve all twenty proteins in one go. Our clotting system has been developed gradually from simpler clotting systems such as those found in invertebrates.

    If you like you can read about the evolution of vertebrate clotting here.

  13. The problem I have in discussing “irreducible complexity” is that we don’t have data we can all agree on to start the discussion. That’s why I find John von Neumann’s thinking on “cellular automata” so helpful. Von Neumann provides an excellent example of truly “irreducible complexity” in the “first self-replicating molecule.” I think everyone has to agree that evolution CAN’T kick in until we have something that is capable of reproducing itself despite minor modifications.

    Salt crystals can reproduce themselves, but without modifications. Mud can mutate without reproducing itself. What von Neumann studied from an information theory standpoint was the minimum size (in bits) of a “program” that could copy itself, even if it had errors.

    Von Neumann came up with a minimum “program size” of 1400 bits for such a “cellular automaton.” Without committing yourself to his conclusions, would you agree that a “cellular automaton” is “irreducibly complex”?


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: