Home » Darwinism, Evolution, Intelligent Design » Playing Devil’s Advocate: Sudden Origins and Irreducible Complexity

Playing Devil’s Advocate: Sudden Origins and Irreducible Complexity

One thing that has always irked me is that rarely on this site do we find any critics of ID attempting to challenge the tools/methods of ID directly. For example, one could claim that “CSI isn’t a reliable indicator of intelligence” or “the explanatory filter breaks down under certain conditions” or “ID regularly produces false positives under x conditions” or “Irreducible Complexity can indeed be overcome via a Direct Pathway” and then show why and/or how. Instead, arguments are almost always made against the implications or we’re arguing over the interpretation of various data. Perhaps these challenges are not made because it’s so difficult to make sustainable arguments in this regard but I’d like to at least see people try. As such, I decided to make a topic on this myself with the last challenge to ID as the subject: “Irreducible Complexity can indeed be overcome via a Direct Pathway”

Behe has claimed there isn’t an evolutionary mechanism that can hope to overcome the IC information barrier via Direct Pathways. The possibility of Indirect Pathways does nothing to change the fact that forming a complex machine via a Direct Mechanism like RM+NS appears impossible. The role of Direct Pathways are, by definition, what makes the bacterial flagellum IC in the first place. The existence of possible Indirect Pathways does not change the designation of IC. But what if another evolutionary mechanism besides the modern synthesis aka Neo-Darwinism/RM+NS is capable of forming an IC structure Directly?

A quick refresher on Jeffrey Schwartz’s Sudden Origins:

The mechanism, the authors explain, is this: Environmental upheaval causes genes to mutate, and those altered genes remain in a recessive state, spreading silently through the population until offspring appear with two copies of the new mutation and change suddenly, seemingly appearing out of thin air. Those changes may be significant and beneficial (like teeth or limbs) or, more likely, kill the organism.

First off, I’ll admit I have not read Schwartz’s book on the subject so I may not do a good job of representing it. As an overall concept it seems to me that these “genes in a recessive state” are like a temporary buffer. In every generation changes are made to this buffer via random mutations, but without harming the survivability of the organisms. If this buffer is implemented before it’s a functional design Natural Selection will kill the organism off. Eventually it’s claimed the mutations in this buffer will result in a successful design and then will be merged with the rest of the organisms. The key to a Direct Pathway is that the evolving mechanisms keep the same function during each revision. Sudden Origins avoids this problem by making it so the function is only required to be active in the final form. By this manner Sudden Origins produces IC Directly.

So unless I’m missing something (besides the obvious possibility–read: I don’t know–of reality not meshing with the details of Sudden Origins) Schwartz’s work in its basic form appears to at least offer a “conceptual challenge” to Behe’s Irreducible Complexity. I myself heavily doubt the validity of Sudden Origins but it would be interesting to point out that it might be a “solution” to the IC Problem which at least is a far better answer than clinging to implausible Indirect Pathways. Schwartz’s own theo…er, posited idea seems to rely far too much on Homeobox genes to do its magic but the concept of a separate buffer where new design plans can be tested and modified without harming the survivability of the species is an interesting one.

Here is a book review of Sudden Origins:

http://www.macrodevelopment.org/library/Schwartz.html

Now that I’m done playing Devil’s Advocate, let me review at a high level the basics of how Homeobox genes work. Homeobox genes determine which genes are expressed and which are not during genetic development. They basically give instructions to build certain components in specific coordinates. Thus, mutations in Homeobox genes can be cause large scale changes. They are also very similar throughout all life. The Pax-6 regulatory group–which is about 130 amino acids long–shares a 94 % similarity between humans and insects. Zebra fish and humans are even closer at 97 percent.

To give a more common example, Homeobox genes resemble a construction site foreman who orders groups of workers to build various building structures in a particular location. However, it’s important to note that this particular foreman doesn’t know how to build walls or windows himself. He just knows how to give orders and couldn’t wield a hammer to save his life. Despite this, the different workers know exactly how to construct the particulars of the items that they are instructed to build. Some workers know how to build concrete walls and others brick walls, windows, archways, etc. In laymen’s terms, without these specialty workers, our foreman is out of a job.

Similiarly, Homeobox regulatory genes work at a high level and don’t get too involved in the details of what they regulate. Even so, Homeobox genes are still extraordinary powerful which is Schwartz relies on them for Sudden Origins. They can “decide” what goes where and how many components get built. But what Homeobox genes cannot is what is truly revelant to this discussion. Homeobox genes do not create *new* genetic information in the traditional sense. Mutations in these regulatory gene sets can cause biological components to not be built (an animal losing their hind legs). They can result in more than the correct number of elements being built (as in the case of Hox-4.6 in chickens which create an extra “thumb”). They can even result in the construction of components in the wrong places. Ultimately, manipulations to these genes can only result in the rearrangement of elements already present in the biological development plan for a given organism.

Homeobox genes reveal complexity, structure and a hierarchical approach to biological development. All of which speak of designer re-use. Any engineer worth his degree realizes the importance and benefit of well organized subsystems. Indeed, these genes represent a level of biological abstraction that Darwinism did not predict. Not to mention, how do you gradually evolve a regulatory gene in the first place? What value does a regulatory gene serve without the genes that know how to “build” components? And if the genes that “build” components existed prior to the appearance of regulatory genes then what selective short term advantage by itself would the ability to regulate other genes have?

A couple weeks back a critic of ID posed the case of humans born with “tails”-a couple extra vertebrae in the spine. I wouldn’t be surprised if these incidents could be traced back to mutations in homeobox genes. But instances of finite macro-level changes do little to advance their case against ID.

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

49 Responses to Playing Devil’s Advocate: Sudden Origins and Irreducible Complexity

  1. Since you decided to focus on the “direct pathway” portion, so will I. And I really only need one reference here.

    http://pharyngula.org/index/we....._grauniad/

    Just ignore the rest of the post and scroll down to the part with the diagrams. Read all of that. Let’s discuss it.

    Myers comes out the gate with a false analogy comparing the complexity in a cell with a potato chip that resembles the face of Jesus. Is that supposed to impress anyone? All it tells me is Myers is an anti-religious ass with a lame argument.

    Then the very next thing he does is says that cellular structures only look like machines implying they aren’t actually machines. This only demonstrates that Myers doesn’t know the definition of a machine. There are real machines in living cells. Really really complex machines.

    I haven’t a clue what his point is with anything further. He thinks he’s demonstrated with a few idiotically simple flow charts how complex machinery self-organizes? Uh no. It’s going to take a bit more than that. Myers here only continues to demonstrate either his ignorance (he doesn’t even know what a machine is), his stupidity (he equates cellular machinery with a potato chip), his dishonesty (he probably does know that his analogy was false and the definition of a machine), or his wickedness but I’d rather not consider that. -ds

  2. First of all, it’s bad form to respond by editing your response into other people’s posts. I speak from experience: I am a staff member at a very active message board.

    On to the main point, you didn’t even bother to respond to the part of the post that I said to read. Patrick’s original post says:

    “Behe has claimed there isn’t an evolutionary mechanism that can hope to overcome the IC information barrier via Direct Pathways.”

    PZ Myers showed how it can happen. How would you go about disproving that such a mechanism works as he says it does?


    First of all this is great form. It reduces clutter. But I’ll offer a fair compromise nonetheless. When we’re on YOUR message board we can do it YOUR way.

    Myer’s idiotically simple flow charts aren’t applicable to machines. Take a simple power transmission device consisting of two gears and a chain. Gear1—-Chain—-Gear2. It’s irreducible. According to Myers logic we can modify it into a more complex irreducible system by first adding a redundant chain so that we can break the first chain without disabling the device. Unfortunately gears don’t work that way. They’ll only accomodate one chain at a time. The flagellar motor is the same way. You can’t just add a redundant drive shaft so you can monkey around with the original without breaking it. There is no accomodation for adding a redundant drive shaft. Myers example fails when applied to machines. But this is hardly surprising since he earlier demonstrates he doesn’t know what a machine is. -ds

  3. With all due respect to Dr. Behe, I’m more impressed with CSI as an indicator of intelligent design than IC.

  4. It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, Instead of theories to suit facts.
    – Sherlock Holmes, “A Scandal in Bohemia”

  5. “One thing that has always irked me is that rarely on this site do we find any critics of ID attempting to challenge the tools/methods of ID directly.”

    The current administrators of this blog don’t allow crititcism of ID, it’s grounds for immediate banning. Back off on the overzealous editing of comments and you’d see plenty of rational, well-informed criticism.

    Ok son. Show us what you got. -ds

  6. crandaddy:
    “With all due respect to Dr. Behe, I’m more impressed with CSI as an indicator of intelligent design than IC.”

    Umm IC is a specialized form of CSI/ SC:

    page 289 of “No Free Lunch”:

    “I want therefore in this section to show how irreducible complexity is a special case of specified complexity, and in particular I want to sketch how one calculates the relevant probabilities needed to eliminate chance and infer design for such sysytems.”

    Then Sir Wm. launches into the DCO computation (discrete combinatorial object) which require parts to be originated, those parts to be localized and those parts to be properly configured. All of which also require information.

    Pdco= Porig x Plocal x Pconfig

    Also what PZ should know- just about anything “works” on paper. However it is the real world that science is concerned with.

  7. -ds,
    Are you deliberately misconstruing Myers? He is not claiming to give a detailed history of anything: only illustrating an abstract point that “irreducible complexity” is not logically equivalent to “unevolvable”. Granted, the argument is a bit abstract but so is much of Dembski’s stuff. Once that point is made, then the argument “it is IC and therefore could not have evolved” – a catagorical statement – becomes “its IC and i cann’t see how it evolved”, which is a much softer statement and maybe says more about the lack of imagination or expertise of the questioner than it does about evolutionary theory. Yes, it poses a challenge for evolutionary scientists but one that they are used to.

    If I was miscontruing Myers you’re free to show how I did it. Myers rebuttal was simplistic, idiotic hogwash that works in the abstract but falls apart applied to the design of actual machines like the flagellum. -ds

  8. “The current administrators of this blog don’t allow crititcism of ID”

    Balderdash. We don’t tolerate tired vitriolic claptrap. Present a good argument against a tenet of ID and we’ll happily engage you.

    http://www.uncommondescent.com.....ock-in-it/

  9. To insouciant,

    IC does NOT mean “it couldn’t have evolved”. The debate is about the mechanism. IOW was the mechanism some blind, purposeless mechanism? Or was it something directed either by the organism (population) or the designed genetic algorithm that provided a solution to the problem- the “problem” being how to achieve motility because the food ain’t coming here anymore.

    Here is a challenge for evolutionary biologists- demonstrate what makes an organism what it is.

    To Scott,

    What happens when the ONLY arguments “against” ID are of the “vitrolic claptrap” type? ;)

  10. -ds said: “Myers rebuttal was simplistic, idiotic hogwash…”

    Very well, let me ask a question: do you all believe that “irreducibly complex” is equivalent to “unevolvable”? Do you believe it means “could have evolved, but I don’t see how” or … what? What exactly is the significance of IC ?

    If truly irreducibly complex it means it couldn’t evolve by natural selection. This was the challenge given by St. Charles hisself back when his theory was actually scientific, before he was sainted, and before his theory of natural selection was transformed into chance worshipping dogma. If something is claimed to have “evolved” the onus of showing how is on the claimer. Science is about demonstration. “Darwin of the gaps” is basically what’s being claimed by the chance worshippers – anything they can’t explain (a gap) is conveniently filled in with an all powerful mechanism random mutation & natural selection. “Darwin of the gaps” is equivalent to “God of the gaps”. The only difference is in which particular deity gets the credit by default. -ds

  11. I narrowed it down a bit, eh Joseph? ;)

  12. Joseph,
    “The debate is about the mechanism” – of course, that is what evolutionary biologists spend their days talking about as well. But i am confused by your terminology. Does “directed either by the organism” mean that it purposefully changes itself somehow? And when you say “the designed genetic algorithm” is that just mean that the thing was “designed”? I’m confused because “genetic algorithms” are something from computer science.

    DNA is (at a minimum) a digital program code controlling a robotic protein assembler. There are mobile elements hopping around like grease in a hot skillet. Epigenetic information abounds. There’s no physical reason why cells can’t have computational capabilities. All the required hardware is there and they have a digital programming language. Vast computational ability is available to nanoscale hardware. Simple mechanical things like levers composed of a small number of atoms can be memory devices able to change state in femtoseconds. Quantum computing devices are even more amazing in their ability to pack computational power into a small space. So far nature appears to have anticipated just about everything human engineers have done so I see no reason to deny her nanoscale computers in living cells. Genetic algorithm takes on a whole new meaning in that context.-ds

  13. insouciant,

    It means that a molecular machine with multiple interlocking components which demonstrate a specified purpose, and whose core will not function unless said components are in place simultaneously, could not have been built by an unguided, purposeless and gradualistic mechanism.

    More:
    http://www.designinference.com.....isited.pdf

  14. “If truly irreducibly complex it means it couldn’t evolve by natural selection.”

    But isn’t that exactly the point of Myer’s example: he presents something which satisfies the definition of IC and then shows that the fact it is IC would not be an obstacle to it’s having evolved. No, it is not a real example, but in logical argument, any counter example of a catagorical statement will do. He seems to prove that IC is not “unevolvable”.

    I showed his example doesn’t work in mechanical devices. Substitue drive gear, chain, and sprocket for A, B, and C in his example and it falls apart because in the real world of machinery you can’t just throw a redundant chain into the machine so you can play with the design of the original chain without breaking the machine. The ability to have redundant parts has to be designed into the machine. Myers evidently knows nothing of machines or design thereof. Either that or he’s lying, delusional, insane, or wicked but I’d rather not consider wicked. He’s off in some theoretical lala land imagining things that can’t possibly be realized in working machinery. -ds

  15. Can’t something that is IC evolve, in part, by drift as well ? Not just natural selection ?

    Of course. That’s what Myers is saying. By introducing a redundant component it removes the selection pressure so random changes can occur in one of the redundant components without losing the original functionality. Unfortunately his abstract example breaks down because you can’t accidently introduce redundant components just any old place in functional machinery. If it wasn’t designed from the gitgo to accept redundant parts then it breaks if you simply try adding a extra identical part. Take for example a drive train in a car. You can’t get redundancy in the power train simply by installing a second engine identical to the first. There’s no room for a second engine. There’s no provision in the rest of the power train to have two engines operating simultaneously. Similarly you can’t just add a second transmission, a second driveshaft, a second set of wheels, etc. Machines don’t work that way. Myers example is abstract hogwash that doesn’t apply to real world situations of irreducibly complex machines. Myers doesn’t understand how machines are designed. I do. That’s my job and I’m quite good at it. -ds

  16. Defenders of ID:
    Don’t forget that IC precludes evolution via natural selection via direct pathways.
    In principle, the immensely improbable evolution via an indirect Darwinian pathway is not ruled out solely by existence of irreducible complexity.

  17. Scott,
    I have in fact read Dembski’s manuscript on IC and it is one reason I am troubled by the concept. First, he gives a definition on page 2 “A functional system is IC if it contains a multipart subsystem that cannot be simplified without destriying the system’s basic function”. No mention of molecular machines in the definition of IC. Meyer’s example seems to be IC according to this definition.

    On page 7, he says “The problem is that for an IC system, its basic function is attained only when all components from the irreducible core are in place simultaneously”. The final stage from Myer’s example seems to satisfy this.

    On page 15, “The logical point is that IC renders biological structures provably inaccessible to Darwinian pathways”. No mention of molecular machines – D. claims the concept applies to ALL biological structures. And he claims it rules out “pathways” which is a general concept. Myers seems to show a schematically how a particular IC structure (a chemical reaction – and all biology is built on chemical reactions) could have evolved.

    I am not really clear on the precise definition of “direct” and “indirect” pathways. Myers claims his example is a “direct” pathway, but maybe Dembski would term it indirect. If it is indirect, what would Dembski say? On page 17, Dembski concedes that indirect pathways are possible but says ” Indirect Darwinian pathways, by contast, are so open ended that there is no way to test them scientifically unless they are carefully specified …”.

    So as far as I can tell, Dembski says that IC rules out evolution along “direct pathways” (which I take to mean that an IC structure cannot be found by following a greedy optimization algorithm) but that they could evolve via Darwinian mechanisms along indirect pathways. I personally find nothing objectionable to any of this but I don’t see how it helps ID. In the end, he just seems to be saying “I will only believe X evolved if you can show me the steps in sufficient detail – and in the mean time, I prefer to believe it was designed by … something”. So what? It still just seems to come down to “i don’t believe it!”.

    And on the other hand, suppose someone in the future came up with a blow-by-blow account of the bacterial flagellum (which Dembski allows as a logical possibility). What would ID supporters do? My guess is just pick something else out and make the same claims. That doesn’t seem very “scientific” to me.

    My emphasis above. What do ID critics do? They create strawmen like telling us what we would do in the future then telling us that their account of our future actions isn’t very scientific. As long as we’re making crap up let’s suppose someone in the future finds God signed his work by leaving a gene in the human genome that constructs a protein that folds into the Hebrew letters “YHWH”. What would chance worshipping neodarwinists do then? My guess is they’d start believing in aliens from outer space that visited the earth and planted the gene there to fool us into thinking that God exists. LOL. I gave you a chance to be critical and when you get backed into a corner you resort to silly strawman arguments. -ds

  18. “The current administrators of this blog don’t allow criticism of ID”

    I’m an administrator here and yet here I am offering a possible new argument against IC. Admittedly it’s on the same level of Myer’s argument: only a logical argument. Darwinists can feel free to run with it and see if it works in reality.

    There are three Indirect Pathways raised as possibilites against IC: “parallel direct evolution (coevolution of components), elimination of functional redundancy (“scaffolding,” the loss of once necessary but now unnecessary components) and adoption from a different function (“cooption,” functional shift of components).” All of those are Indirect Pathways. “a fourth route, serial direct evolution (change along a single axis), could not produce multiple-components-required systems.” Note the sections in quotes which I yanked from the PT people.

    If Myer’s example of a Direct Pathway is useful in real life solutions then he should apply it to the bacterial flagellum and produce a detailed account of this Direct Pathway.

    Charlie: “In principle, the immensely improbable evolution via an indirect Darwinian pathway is not ruled out solely by existence of irreducible complexity.”

    Yes, we realize that. We’d also like to see posited Indirect Pathways backed up by experimental evidence.

  19. You’re right, sorry Patrick.
    My reminder, based upon what I thought was transpiring in the comments, was completely redundant.
    Too much surfing too fast – I forgot that the initial post had already addressed the distinction.

  20. I’d like to get some feedback on my logical argument against IC. I spent about a week mulling it over and I did a mini UD “peer review” before posting it publicly but I’d like to see if there any problem with it (at a high level).

  21. Patrick, I too lament the scarcity of claims like those that you mentioned:

    1. “CSI isn’t a reliable indicator of intelligence”
    2. “the explanatory filter breaks down under certain conditions”
    3. “ID regularly produces false positives under x conditions”

    To remedy this, I’ll gladly make a concise and hopefully well-reasoned case for any of the above three claims. You can pick which one you’d like to see.

  22. Charlie:
    “Don’t forget that IC precludes evolution via natural selection via direct pathways.
    In principle, the immensely improbable evolution via an indirect Darwinian pathway is not ruled out solely by existence of irreducible complexity. ”

    Natural selection is a result of differences in survival and reproduction…

    We cannot forget that the mutations are allgedly random, genetic accidents.

    And “genetic algorithms” are something from computer science. However they may give us some insight to the inner “workings” of biological information:

    [I]DNA and the Origin of Life: Information, Specification, and Explanation[/I] pg 233-85 (including a dozen pages worth of references) of [I]Darwinism, Design and Public Education[/I]

    pg 237
    [quote]
    Rather, the coding regions of DNA function in much the same way as a software program or machine code, directing operations within a complex material system via highly complex yet specified sequences of characters. As Richard Dawkins has noted, “The machine code of the genes is uncannily computer-like.” Or as software developer Bill Gates noted, “DNA is like a computer program, but far, far more advanced than any software we’ve created.” Just as the specific arrangement of two symbols (0 and 1) in a software program can perform a function within a machine environment, so, too, can the precise sequencing of the four nucleotide bases in DNA perform a function within the cell.[/quote]

  23. Patrick wrote:
    “I’d like to get some feedback on my logical argument against IC.”

    Hi Patrick,

    This seems to be the core of your argument:
    “As an overall concept it seems to me that these “genes in a recessive state” are like a temporary buffer. In every generation changes are made to this buffer via random mutations, but without harming the survivability of the organisms. If this buffer is implemented before it’s a functional design Natural Selection will kill the organism off. Eventually it’s claimed the mutations in this buffer will result in a successful design and then will be merged with the rest of the organisms.”

    I see two issues with this idea:

    1. Since the mutations aren’t being expressed as they accumulate, they aren’t being filtered by natural selection. “Bad” mutations are just as likely to survive in the buffer as “good” ones. The odds against randomly accumulating a set of “good” mutations that can act in concert to produce a useful, irreducibly complex structure seem pretty steep to me. (In other words, there are staggeringly more non-functional combinations of random mutations than functional ones, much less IC ones, so the odds of hitting an IC combination are low. Taking the mutation rate into account, you simply wouldn’t have enough time.)

    2. How does the organism “know” when the buffer contains a successful design and can be “switched on”? What mechanism tests the buffer contents for viability?

    Regards,
    Valerie

  24. “suppose someone in the future came up with a blow-by-blow account of the bacterial flagellum (which Dembski allows as a logical possibility). What would ID supporters do? My guess is just pick something else out and make the same claims. That doesn’t seem very “scientific” to me.”

    insouciant, out of all the various Indirect Pathways in the search space there WILL BE some possibilities that are a “a blow-by-blow account of the bacterial flagellum”. The point is to not show these possibilities exist; we KNOW they exist. The point is to show they can and do occur in reality (even if it’s only in the controlled environment of a lab).

    ” Indirect Darwinian pathways, by contrast, are so open ended that there is no way to test them scientifically unless they are carefully specified …”

    Using the mere existence of Indirect Pathways as a general response isn’t an answer. We know they exist. Pick a SPECIFIC Indirect Pathway and test it. That’s Dembski’s point.

  25. And Dembski seems to make it clear that the indirect pathways are so rediculously improbable that they are in effect, impossible.

  26. Valerie,

    The weaknesses of this argument obviously are the weaknesses inherent to the details of Sudden Origins.

    1. I agree that’s the largest weakness with Sudden Origins. I probably should have pointed that out in my article along with my discussion on Homeobox genes. Of course, the problem with other models is that while “bad combinations” are filtered out fairly quickly it’s still a matter of scrambling through the search space looking for a “good combination”.

    2. This is what Schwartz says (paraphrased by a reviewer) about the the implementation of the buffer itself:

    “two mates often find each other that are heterozygous for this mutation. One quarter of the offspring then have two recessive alleles and the trait is expressed. The phenotype of these individuals is so different that they prefer each other for mating. Then, somehow, the mutated gene is converted to the dominant state.”

    Viability is tested via natural selection; i.e. if the design isn’t viable the organism dies. But this doesn’t kill off the buffer; it still exists in the rest of the population. Problem is that a “good combination” might come about and the trait is never expressed before the buffer is shuffled into yet another “bad combination”.

  27. Please correct me if I am wrong – a system is IC if, when one component is removed, the system fails to function. And Behe claims IC can’t evolve in a stepwise manner.

    If that is the definition, whether the system is a flagellum or a simple circuit as shown by Myers, is not relevant to whether or that system is IC. The node in the circuit that Myers demonstrated is EITHER IC or NOT, based on the definition provided by Behe (restated by me here – correct me if I am wrong). The fact that this circuit is not a complex machine or a flagellum has no logical bearing on whether the circuit, as shown by Myers, is IC or not.

    Secondly, DS, you claim that the way machines are designed doesn’t allow for tolerance to the incorporation of redundancy.

    Quote: “If it wasn’t designed from the gitgo to accept redundant parts then it breaks if you simply try adding a extra identical part. Take for example a drive train in a car. You can’t get redundancy in the power train simply by installing a second engine identical to the first. There’s no room for a second engine. There’s no provision in the rest of the power train to have two engines operating simultaneously.”

    Well, biological organisms can withstand plenty of redundancy. Your machine metaphor doesn’t work with biological systems. Gene duplications happen all the time without deleterious consqequences.

  28. senatorchunk -

    A system with built-in redundancy is by definition NOT IC since one could remove one of the redundant parts without impacting functionality. Generalizing that machine metaphors don’t work based on that logic seems flawed since the metaphor applies to either IC or non-IC systems. A machine is a machine whether it is IC or not.

  29. Dougmoran,

    I think you misunderstand. On the WAY to IC there may be redundancy. But over time, this redundancy may go away and IC may evolve. That is what Myers demonstrated with his circut model.

    DS seemed to say that machines can’t tolerate redundancy, therefore, Myers argument was a false analogy.

    But biological systems can tolerate redundancy.

  30. senatorchunk, my example is in the same boat as Myers’s. They’re fine as abstract logical arguments but when applied to reality they “appear” to quickly fall apart. If that weren’t the case then why hasn’t Myers produced an account of the Direct Pathway for the flagellum? While I wouldn’t phrase it quite so…um, in-your-face that’s why DaveScot started talking about “theoretical lala land”.

  31. “I agree that’s the largest weakness with Sudden Origins. I probably should have pointed that out in my article along with my discussion on Homeobox genes. Of course, the problem with other models is that while “bad combinations” are filtered out fairly quickly it’s still a matter of scrambling through the search space looking for a “good combination”.

    Patrick,
    It’s true that natural selection still has to “scramble through the search space”, but I think you’d agree that the scrambling is easier by orders of magnitude if bad combinations are being filtered out as you go. Where we will likely disagree is on how easy it becomes, and whether that is easy enough to make it a plausible mechanism for macroevolutionary adaptation.

    The thing about the Schwartz paraphrase is that he seems to be talking about “turning on” one mutation at a time:

    “two mates often find each other that are heterozygous for this mutation. One quarter of the offspring then have two recessive alleles and the trait is expressed.”

    But surely you believe that an IC system would require more than one mutation. Otherwise, you’d essentially be admitting that ordinary natural selection could produce IC.

    On the other hand, if an IC structure required multiple mutations, you’d need a mechanism for turning them all on simultaneously and also for converting them to dominance simultaneously.

    Regards,
    Valerie

  32. Umm, I don’t think homeobox genes can help with the bacterial flagellum.

    IC from “No Free Lunch:

    IC- A system performing a given basic function is irreducibly complex if it includes a set of well-matched, mutually interacting, non-arbitrarily individuated parts such that each part in the set is indispensable to maintaining the system’s basic, and therefore original, function. The set of these indispensable parts is known as the irreducible core of the system.

    Numerous and Diverse Parts If the irreducible core of an IC system consists of one or only a few parts, there may be no insuperable obstacle to the Darwinian mechanism explaining how that system arose in one fell swoop. But as the number of indispensable well-fitted, mutually interacting,, non-arbitrarily individuated parts increases in number & diversity, there is no possibility of the Darwinian mechanism achieving that system in one fell swoop.

    Minimal Complexity and Function Given an IC system with numerous & diverse parts in its core, the Darwinian mechanism must produce it gradually. But if the system needs to operate at a certain minimal level of function before it can be of any use to the organism & if to achieve that level of function it requires a certain minimal level of complexity already possessed by the irreducible core, the Darwinian mechanism has no functional intermediates to exploit.

  33. Patrick,

    I am not sure what you mean – “when applied to reality they appear to quickly fall apart”.

    Two points.

    1) These ideas are abstract, yes. But they are demonstrations of a logical fact, not mere theoretical fancies. The fact is this: If an IC system is defined: “A system that, when one part is removed, it fails to function” then IC CAN, in principle, evolve. Whether it would be common or not is another question. But in principle, and as defined, IC can evolve.

    2) While the circuit model may appear abstract on paper, there is plenty of evidence for this type of evolution in gene regulatory networks via gene duplication.

    Also, if we use a different definition for IC (no longer using the “if you remove one part and it fails” definition), but, instead, use a definition such as “A system that, due to it’s fundamental interconnectedness, CAN NOT evolve in a stepwise manner”, then the presence of IC systems in nature, whether direct or indirectly evolved, is a challenge to neo-darwinisim. However, as far as I know, no such system has been identified.

    Finally, your idea of a “temporary buffer”, though perhaps in different terms, is actually an important idea in evolutionary biology, under the term “neutrality”. It is certainly reasonable to argue that many mutations may remain in populations if the mutations have no effect in one environment, but may be favorable in another environment.

  34. Patrick,

    If you argued that the mutations were side-by-side on a chromosome, then they would be more likely to remain together as a unit, and Schwartz’s mechanism could pair up two instances of the unit, causing it to be expressed all at once.

    In this case the likelihood of success would be determined by the rate of mutation relative to the probability of separation via meiosis. To improve the odds, you could envision a mechanism to keep the genes linked (by preventing crossover from occurring in the area).

    The problem, of course, is that the genes “requiring” mutation are unlikely to be side-by-side or even close, so crossover is likely to separate them before the full ensemble has been assembled.

    In any case, the problem of not filtering out “bad” mutations remains.

    Valerie

  35. “One thing that has always irked me is that rarely on this site do we find any critics of ID attempting to challenge the tools/methods of ID directly.”

    Has anyone come up with a method for actually detecting ID?

    What predictive powers does ID have?

    How can ID be falsified?

    What peer reviewed journals have published ID research papers?

    These are basic questions that anyone familiar with science need to be asking, the questions need to be answered by the ID camp if they ever want to be taken seriously.

    Dembski has come up with a method in his book No Free Lunch. ID predicts you can culture bacteria forever and not get a nucleus. ID can be falsified by verifying in a lab that CSI like that found in a prokaryote ribosome can evolve without intelligent agency. When an editor let an ID symptathetic article get published in a mainstream journal the NCSE and Smithsonian conspired to ruin his career (Rick Sternberg). There are some articles by John Davison at the right sidebar that are ID symptathetic and appear in peer reviewed journals. When he published the first one in 1984 his career was also ruined. Thanks for playing. But I thought you were here to criticize not ask basic questions. -ds

  36. I, for one, am disappointed that nobody took secondclass up on his offer. I would have chosen option 1: The case for “CSI isn’t a reliable indicator of intelligence.”

    Just for that, I’ll write an article describing CSI using my lottery analogy. -ds

  37. “ID predicts you can culture bacteria forever and not get a nucleus.” I thought the idea behind ID was that the designer had somehow included all the required genetic variation for complexity from the word go. So unless this genetic variation has been removed then (according to ID) if you culture bacteria forever then you might well get a nucleus because you will allow that complexity to emerge again.

    The eukaryote nucleus exhibits specified complexity and modern bacteria do not appear to have any way of generating the specified complexity (insufficient probabilistic resources). A design inference is thus warranted. You ask for predictions then say they aren’t our predictions? What kind of crap is that? I have a sign hanging in the main salon of my boat which reads

    When we’re on YOUR boat we’ll do it YOUR way!

    It applies to theories too. -ds

  38. And…

    “ID can be falsified by verifying in a lab that CSI like that found in a prokaryote ribosome can evolve without intelligent agency.”

    If this were to happen (and actually it would take far too long to be practical) how would you know that an intelligent agency had not been involved? Unless you specify the mechanism by which the intelligent agent operates then it is always possible to claim the variation was directed by some external force towards an end and not truly random.

    When we’re talking about YOUR theory we’ll falsify it YOUR way.

    Science can never confirm or deny supernatural causes. How do you know that the the universe isn’t 1 second old and we weren’t all placed here by supernatural agency complete with memories and convincing evidence that everything was here more than 1 second ago?

    If it happens and there’s no evidence of intelligent agency acting through known natural laws we’ll consider it scientifically falsified. What’s the problem pal, afraid to let a falsifiable theory that makes predictions compete on a level playing field with your precious random mutation + natural selection narrative? It sure looks that way from here. I’d be afraid too if I were you. RM+NS is a lame duck when it comes to explaining the origin of novel cell types, tissue types, organs, and body plans. You know that, don’t you? It’s okay. It’ll only hurt for a minute if you admit it. :-) -ds

  39. SailorMon asks:
    Has anyone come up with a method for actually detecting ID?

    Yes. Scientists and non-scientists alike detect design daily.

    What predictive powers does ID have?

    “The Privileged Planet” makes several predictions specific to their premise. However ID would predict that we would observe/ find IC and CSI and that neither can arise via unintelligent, blind/ undirected processes.

    How can ID be falsified?

    By demonstrating that CSI and IC can arise without an intelligent agency. IOW demonstrate that life can arise from non-living matter via some blind watchmaker-type process and you will have fulfilled Behe’s and Dembski’s criteria. Observe:

    As Dr. Behe wrote:
    “Coyne’s conclusion that design is unfalsifiable, however, seems to be at odds with the arguments of other reviewers of my book. Clearly, Russell Doolittle (Doolittle 1997), Kenneth Miller (Miller 1999), and others have advanced scientific arguments aimed at falsifying ID. (See my articles on blood clotting and the “acid test” on this web site.) If the results with knock-out mice (Bugge et al. 1996) had been as Doolittle first thought, or if Barry Hall’s work (Hall 1999) had indeed shown what Miller implied, then they correctly believed my claims about irreducible complexity would have suffered quite a blow. And since my claim for intelligent design requires that no unintelligent process be sufficient to produce such irreducibly complex systems, then the plausibility of ID would suffer enormously. Other scientists, including those on the National Academy of Science’s Steering Committee on Science and Creationism, in commenting on my book have also pointed to physical evidence (such as the similar structures of hemoglobin and myoglobin) which they think shows that irreducibly complex biochemical systems can be produced by natural selection: “However, structures and processes that are claimed to be ‘irreducibly’ complex typically are not on closer inspection.” (National Academy of Sciences 1999, p. 22)

    Now, one can’t have it both ways. One can’t say both that ID is unfalsifiable (or untestable) and that there is evidence against it. Either it is unfalsifiable and floats serenely beyond experimental reproach, or it can be criticized on the basis of our observations and is therefore testable. The fact that critical reviewers advance scientific arguments against ID (whether successfully or not) shows that intelligent design is indeed falsifiable.

    In fact, my argument for intelligent design is open to direct experimental rebuttal Here is a thought experiment that makes the point clear. In Darwin’s Black Box (Behe 1996) I claimed that the bacterial flagellum was irreducibly complex and so required deliberate intelligent design. The flip side of this claim is that the flagellum can’t be produced by natural selection acting on random mutation, or any other unintelligent process. To falsify such a claim, a scientist could go into the laboratory, place a bacterial species lacking a flagellum under some selective pressure (for mobility, say), grow it for ten thousand generations, and see if a flagellum–or any equally complex system–was produced. If that happened, my claims would be neatly disproven.

    How about Professor Coyne’s concern that, if one system were shown to be the result of natural selection, proponents of ID could just claim that some other system was designed? I think the objection has little force. If natural selection were shown to be capable of producing a system of a certain degree of complexity, then the assumption would be that it could produce any other system of an equal or lesser degree of complexity. If Coyne demonstrated that the flagellum (which requires approximately forty gene products) could be produced by selection, I would be rather foolish to then assert that the blood clotting system (which consists of about twenty proteins) required intelligent design.”

    What peer reviewed journals have published ID research papers?

    This question misses the point entirely. What is “ID research”? ID is about conducting scientific research and being allowed to come to a design inference if the data affords it. After all if one or more IDists went into a lab and genetically engineered a bacterial flagellum, would that mean that all flagellum were designed? No.

  40. “ID predicts you can culture bacteria forever and not get a nucleus.”

    Does this mean that a nucleus will not evolve? Or does it mean that the designer will not and/or can no longer design a nucleus? IF A NUCLEUS DOESN’T APPEAR, MIGHT IT NOT MEAN THAT THE DESIGNER IS SOMEHOW PREVENTING IT FROM EVOLVING? How could you tell? Besides, how does one culture bacteria forever and THEN look at the results?

  41. secondclass,

    Go ahead and pick one. Try to make it a unique argument and not a copy and paste job from various anti-ID sites/books.

    valerie and others,

    I knew from the get-go that my argument would fall apart quickly when exposed to details. I was primarily interested in seeing if anyone could spot any high level flaws in the argument. I take it no one sees any major flaws since you’re skipping straight to the details. Really, the only way I see to save this argument is if organisms are capable of self-modification to a certain degree; meaning that this high concept buffer is moderated and checked for extremes. Which really amounts to separating this argument from Sudden Origins (which doesn’t appear sustainable) and the creation of a new hypothesis.

    Anyway, as I said before anyone can feel free to take the basic idea of this argument and run with it.

  42. Re 38 – (ignoring the vitriol) I didn’t mention the supernatural. All I wanted to say was that if you observed a complex function evolve how you would you know that there was no designer involved? One example of how a designer might do this would be through influencing the variation part of RM+NS – so it becomes directed variation + NS. This doesn’t strike me as particulary supernatural – the designer has got to implement its design somehow – but if you don’t like it just tell me how the designer does implement its designs and then we can observe if it is happening or not.

    All I wanted to say was that if you observed a complex function evolve how you would you know that there was no designer involved?

    You cannot ever rule out unknown causes in science. This is why all science is tentative. However, before presuming a designer interfered there must be some observation to warrant it. A supernatural designer who by definition cannot be observed is out of bounds as an explanatory mechanism. If no observable cause can be identified then the causation must be left as an unexplained phenomenon. If a design inference is warranted but a designer has not been observed then the nature of the designer and the mechanism used to impliment the design must remain unexplained, but that doesn’t negate the fact that a design inference is warranted. Sometimes science leads to questions that are difficult or impossible to answer. This may be one of those. -ds

  43. Re 42. OK. I understand the argument. I cannot conceive how you are going to observe this designer and it seems like you can’t either. I thought one possibility might be that the designer was influencing the outcome of meiosis at the quantum level – at least it is a mechanism – but you consider this supernatural so I will leave it.

    How could that happen whilst obeying known laws of physics? I’m a materialist. If there’s no scientific explanation then it goes off my radar screen. One can make a design inference based on empirical evidence, logic, and math. Nothing more is revealed beyond the inference. A mechanism is not revealed and a perpetrator is not revealed. However, as a materialist I’m confident that if no material mechanism is in sight that will evolve a nucleus in a modern bacteria then we will not observe a nucleus emerging in a modern bacteria. Presumably if we do observe it we will see it happen in a stepwise neoDarwinian fashion through mutation and selection where each step makes sense without having to postulate the existence of an unobserved operator influencing events at the quantum level. At that point I will concede the design inference was wrong. -ds

    The consequence of avoiding the mechanism for implementation is that you can only argue for ID by the success or failure of Darwinian explanations. And this is the essence of why ID is not science. The whole of the evidence for ID is reduced to:

    We cannot find a plausible Darwinian explanation for the appearance of phenomemon A therefore phenomenon A was designed.

    BZZZZZZZZZZZZZZZZZZT. Wrong. I have taken some time to explain to you how specified complexity warrants a design inference. You just blew it off like I didn’t write anything. Part of the inference is making a determination of probabilistic resources. The principle probabilistic resource known to us is mutation + selection so of course the focus is whether that is sufficient. The same claim to an argument from ignorance can be made of neoDarwinism’s claims. No one observed a nucleus evolving. The entire neoDarwinian story of long past events is based on the claim that if mutation + selection was not responsible then what else could it be? You’re in denial if you think there is more strength in the neoDarwinian claim.

    The problems with this are:

    1) A lot of phenomena are explicable through RM+NS and it always possible that any other phenomenon can be explained at some stage.

    A lot of phenomenon can be explained by intelligent agents tinkering with genomes too. We have observed exactly the same number of intelligent agents working in the remote past as we’ve observed random mutation + natural selection working in the remote past. That would be zero observations for each. Both are known quantities in the present.

    2) The detail of much of the mechanism for RM+NS is well known and regularly being reinforced (with the occasional discovery which gives pause for thought). Think back to how little of the mechanism Darwin knew and yet we have discovered so many things that support his theory and show how it works (meiosis, particulate inheritance, mutation of DNA to name a few)

    Each of these fits just as well into a front-loaded theory of evolution. The front-loaded theory explains everything and explains it better than neoDarwinian theory. The only gap in the front loaded theory is where the first cell came from but in a universe as old and large as this one it’s not difficult to suppose that intelligent life arose elsewhere first or indeed maybe intelligence came first and the observable universe came next. Who knows. In any case the origin of the first cell is beyond the scope of either ID or neoDarwinian evolution.

    3) We have evidence that the complex phenomena that are hard to explain through RM+NS took a very long time indeed and so it would be extremely hard to observe them and even harder to observe them not happening

    So in the absence of observation and experiment because “these things take too long to happen” you get to fabricate an elaborate history of how life evolved by accident and have your story legally enshrined as the only story that science may offer? I don’t think so. Darwin of the Gaps is pseudoscience – a security blanket for the intellectual fulfillment of atheists as Richard Dawkins so aptly put it.

    4) Even if we were forced to reconsider RM+NS through some future evidence it does not follow that there is a design. There may be some other explanation than RM+NS (or a refinement of RM+NS) which is not designed – e.g. perhaps the mechanism of variation is far more limited than we realise forcing evolution down certain routes and greatly increasing the chance of those routes.

    Again, you simply deny a design inference like it doesn’t exist. Incredible.

    Anyhow, as with all these discussions, I know I am not going to convert anyone. So I guess that’s it.

    I guess it is. Thanks for playing. Better luck next time.

  44. senatorchunk says:

    The fact is this: If an IC system is defined: “A system that, when one part is removed, it fails to function” then IC CAN, in principle, evolve.

    But of course, we know that isn’t how IC is defined. Behe coined the term, and defined it thusly:

    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.

    So the knockout test is a necessary, but not sufficient condition, to label something as IC. It should be noted that Behe later offered an alternative definition, that allowed one to quantify the level of IC, by noting how many unselected (neutral) mutations would be needed for the system to evolve from a given precursor.

    Maybe somebody can help me with the PZ Myers example, and tell me what I am missing. He asserts that the A->B->B’->C pathway is IC. But it is clearly not. Remove B’, and you are back to the original pathway which supposedly functioned. That pathway is also apparently assumed to be IC, but PZ doesn’t “evolve” it, but provides that as the starting point.

    Am I missing something?

  45. Nevermind, I guess I should have paid more attention to the text and not the letters. B in the final pathway is not really B, but should be B” or some other such designation.

  46. “…when one part is removed, it fails to function”

    “…wherein the removal of any one of the parts causes the system to effectively cease functioning.”

    Roger: These are the same thing.

  47. Once again- for Sparrowhawk:

    IC from “No Free Lunch:

    IC- A system performing a given basic function is irreducibly complex if it includes a set of well-matched, mutually interacting, non-arbitrarily individuated parts such that each part in the set is indispensable to maintaining the system’s basic, and therefore original, function. The set of these indispensable parts is known as the irreducible core of the system.

    Numerous and Diverse Parts If the irreducible core of an IC system consists of one or only a few parts, there may be no insuperable obstacle to the Darwinian mechanism explaining how that system arose in one fell swoop. But as the number of indispensable well-fitted, mutually interacting,, non-arbitrarily individuated parts increases in number & diversity, there is no possibility of the Darwinian mechanism achieving that system in one fell swoop.

    Minimal Complexity and Function Given an IC system with numerous & diverse parts in its core, the Darwinian mechanism must produce it gradually. But if the system needs to operate at a certain minimal level of function before it can be of any use to the organism & if to achieve that level of function it requires a certain minimal level of complexity already possessed by the irreducible core, the Darwinian mechanism has no functional intermediates to exploit.

    Further any IC system is considered to be a discrete combinatorial object (DCO). Every DCO requires three things:

    1) Parts to be created- P orig
    2) Parts to be localized- P local
    3) Parts to be configured= P config

    Therefore a DCO = Plocal x Porig x Pconfig

  48. Sparrowhawk says:

    Roger: These are the same thing.

    I agree. I didn’t say the knockout test was an inaccurate interpretation of what Behe said, I said it was an incomplete definition of IC:

    So the knockout test is a necessary, but not sufficient condition, to label something as IC.

  49. So is anyone going to attempt to challenge the tools/methods of ID directly? I’ve been busy the last couple days and I’m kind of surprised no one has even started yet.

Leave a Reply