KairosFocus’ self-defeating challenge

KairosFocus, he who shall not be real-named (Henceforth KF), habitual censor over at Uncommon Descent, perpetually crows about his long-standing challenge:

provide a 6,000 word feature-length article that justifies the Darwinist tree of life from its OOL roots up through the Cambrian revo — as in Darwin’s Doubt territory — and other major formation of body plans up to and including our own origins, and we will host it here at UD, one of the leading ID blogs in the world. We are perfectly willing to host a parallel post with another site. Only, you must provide thesis and observation based evidence that solidly justifies your conclusions in light of inference to best explanation, the vera causa principle and other basic principles of sound scientific induction. Also, you must actually argue the case in outline, a summing up if you will. You must strive to avoid Lewontin’s a priori evolutionary materialism, and if you would redefine science on such terms you will have to reasonably justify why that is not a question-begging definition, in a way that is historically and philosophically soundly informed. Of course, you may link sources elsewhere, but you must engage the task of providing a coherent, non-question-begging, cogent argument in summary at the level of a feature-length serious magazine article . . . no literature bluffs in short.

[some format lost because I can’t be arsed]

KF is of course free to set the bar for his personal satisfaction at whatever pathetic level of detail he requires, but given that he’s often accused of being a massive hypocrite I’m sure he’ll be happy to provide us with a corresponding ID narrative.

I mean, ID isn’t just a negative case against Evolution, is it? 😉

Things I’m sure he’s eager to include:

Who is / was the designer?
What was their motivation(s)?
What was their method of fabrication?
How many design interventions were there?
What specifically was designed?
What specifically wasn’t designed?

Please feel free to add your questions in the comments.

I’d ask that if math is invoked for any design justification then it is comprehensively completed and not just talked about in a big numbers / hand-wavy sort of way. Any new concepts you bring to the table must be empirically tested rigorously so we can attest to their design detection capabilities.

Thanks in advance KF, we know you’ll engage us in good faith and we’re eager to have productive dialogue.

EDIT: vjtorley has started a parallel thread over at UD. Thanks VJ.

 

Let’s hope we can have some good, scientific dialogue.

150 thoughts on “KairosFocus’ self-defeating challenge

  1. I would appreciate it someone could provide me with Kariosfocus’ email address. I’ve done a google search and haven’t found it. He seems to have a web site, but there’s no contact information.

  2. petrushka:
    Too late. I accepted the challenge and my essay has been posted. Now it’s just a matter of waiting until I’m allowed to participate.

    You’ll probably want to compare the posted-at-UD version of your essay to the version you have stored on your hard drive—and, as well, you’ll probably want to do this comparison repeatedly, say, once a week or once a month or whatever other frequency you feel comfortable with.

  3. petrushka:
    I would appreciate it someone could provide me with Kariosfocus’ email address. I’ve done a google search and haven’t found it. He seems to have a web site, but there’s no contact information.

    PM me at AtBC and I’ll see what I can find 😉

  4. petrushka:
    I would appreciate it someone could provide me with Kariosfocus’ email address. I’ve done a google search and haven’t found it. He seems to have a web site, but there’s no contact information.

    At the risk of inducing a bout of pearl-clutching, scroll to the bottom of the 78,000 word opus that you get when you click on his UD handle. Look for the “Author’s written permission” link…
    Moderators, you have my permission to delete this post once it has served its purpose…

  5. I think it’s fairly obvious from the comments of Joe and KF that they do not want me to post there. And even if they did, I don’t see how KF could do anything. He says he isn’t a moderator.

    It would be OK with me to have my comments relayed from here.

  6. Reciprocating Bill 2:
    I can post your comments to UD, if you like.

    I addressed a couple of questions specifically to VJT and gpuccio. I think there’s less animosity in that direction and it might be more productive.

    1.
    Douglas Axe appears to be a competent lab technician. My question is, would his protocol have found the citrate metabolism sequence found by Lensky’s population?

    I’m interested in whether Axe has experimentally considered multiple enabling mutations.

    If not, of what probative value is his work?

    2.
    As long as they are arguing from analogy and metaphor, I’d like to ask gpuccio what language is spoken in Italy, and what language was spoken in Italy 2000 years ago, and whether the transition was a saltation.

    And while on the subject of language evolution, I’d like to ask him whether Basque was created by the designer to have nothing in common with any other known language.

    3.
    Another analogy:

    I would liken Axe to an investor who is unwilling to accept any temporary losses. What Axe does with protein evolution is argue that an investor cannot make money unless the value his portfolio increases with every trade and with every change in the market.

    No sideways trades allowed, and no temporary losses.

    What Axe does with his experimental work is lose money and argue that because he lost, it is impossible to win.

    If VJT or gpuccio would respond, I’d appreciate it.

  7. I really have to reproduce KF’s latest in full here:

    “EA: I of course cannot speak for UD’s mods. However, when I issued the challenge, P had comment privileges here. I doubt that loss of such privileges — if that is so (is it, UD mods?) — would have been without fairly serious cause. On correction, the response was not actually posted here but at TSZ; VJT who seems to monitor there picked it up and reposted here. There is a parallel discussion, linked from here. Having been there to see how the challenge was made, I found the general context so nastily personal and willfully misleading that I posted some correctives above. I will not try to comment at TSZ so long as it drifts into the heckler’s convention mentality I saw; life is too short for me to waste time and energy dealing with behaviour like that; other than to set the record straight. That said, if you or any other person see responses or remarks there that need a FTR or FYI or point-by-point response on merits or even know where any earlier actual overall response by P or another person is, why not clip-link such here? For that matter, if you have your own thoughts, those too would be welcome. KF”

    So Patrushka, you’re not banned! or if you were, its your fault.

    KF won’t be coming here because of tone, not because he’ll be called out on his no-math.

    So there you have it, we can’t engage there, they won’t come to uncensored venues. The eco-chamber has shut its doors and please don’t point out that CSI is incalculable. Obviously I hold these people, KF especially, in the lowest regard. He’s peddling snake oil, and he knows it. A scientific non-starter and a terrible advert for Christianity.

  8. Hi Petrushka and everyone,

    If you need to contact kairosfocus or send him a manuscript for an essay, you’re welcome to email me. I can contact him for you.

    Regarding the computation of the FSCO/I in a living thing, kairosfocus has already made a ballpark estimate. See here: http://www.uncommondescent.com/intelligent-design/a-world-famous-chemist-tells-the-truth-theres-no-scientist-alive-today-who-understands-macroevolution/#comment-492095

    “For instance over the past year — through publishing Darwin’s Doubt — Meyer has put on the table the issue of the Cambrian fossil life revolution, which operates at the level of phyla and subphyla. The info challenge implied in this can be seen from the quantum of info to provide a credible minimal genome for a unicellular first common ancestor, 100 – 1,000 k bits, half that in 4-state bases. Then, to get to reasonable estimates on body plans, we are looking at 10 – 100 mn bits each, dozens of times over.”

    So there you have it. 10 to 100 million bits for a complex animal, 100,000 to 1 million bits for a bacterium.

    I’ve been trying to come up with a good number for the FSCO/I in an organism for ages now. I came across this paper recently: “A meta-analysis of the genomic and transcriptomic composition of complex life” by Ganqiang Liu et al. in Cell Cycle 12:13, 127-138, July 1, 2013. The article suggests that it is non-coding DNA, rather than proteins, that best corresponds to complexity in animals (since most animals have about the same number of proteins), and that non-coding DNA correlates well with the number of cell types in an organism. Unfortunately, however, there are no figures from which I can deduce the number of bits for a given organism. This paper is also useful: “The relationship between proteome size, structural disorder and organism complexity” by Eva Schad, Peter Tompa and Hedi Hegyi, in Genome Biology 2011, 12:R120, at http://genomebiology.com/2011/12/12/R120. It addresses the G-value paradox, among other things. For a graph showing the number of cell types in various organisms, see the top graph here: http://genomebiology.com/2011/12/12/R120/figure/F2 . As you can see, the number of cell types is about 1 in bacteria, up to 10 in unicellular eukaryotes, 10-30 in plants, and 15 to 170 in animals, with vertebrates having 100+ (150-170 for mammals and birds) and insects coming in second, with 50 to 100. Some plants are more complex than some animals. Contrary to what PZ Myers would have you believe, there does seem to be a correlation between number of cell types and organismal complexity. And the number of cell types in Homo sapiens is 169, not the popularly quoted figure of 210.

    Anyway, here’s another approach to calculating FSCI in organisms that might give a ballpark figure. Let’s say that a protein typically has 500 bits of information. Now consider singleton proteins. Let’s say a typical singleton has 10% dissimilarity to its nearest neighbor. The incremental cost of getting to a singleton protein will therefore be at least 50 bits. Kozulic estimates that each species has about 1,000 of these singleton proteins which distinguish that species from all other species. So the emergence of a new species means an increment of 50,000 bits. Species have a lifetime of about 5 million years, and the evolution of complex animals kicked off about 530 million years ago in the early Cambrian. (I’m disregarding the information amassed in the previous 3.5 billion years as trivial by comparison, as one-celled organisms are far less complex.) 530 divided by 5 is 106, or around 100. So if we confine ourselves to proteins, complex animals must have amassed 50,000 x 100 or 5 million bits of information (assuming they haven’t lost most of the information they’ve acquired). This figure is likely to be the same for all phyla of complex animals. Now consider the fact that only a small portion of the genome codes for proteins. We also have to consider the fact that some of the genome contains junk – although this is likely far less than evolutionists typically assume. Anyway, let’s assume that in vertebrates, the non-protein-coding information in the genome exceed the protein-coding information by a ratio of 10:1. That means that the total amount of FSCI in a vertebrate is around 50 million bits. So that’s my back-of-the-envelope estimate. Is everyone happy now?

  9. I posted yesterday a link to the original KF challenge thread at TSZ. The reason we were here is because most of us had already been banned.

    So KF’s statement is factually incorrect. I had been banned for my response to Barry’s question.

  10. VJT: How do you know how many bits of information are in a protein?

    If you start with a random sequence that happens to code for a binding function, how many bits of information does it have?

    Assuming it undergoes “microevolution” and becomes a more efficient molecule, how do you count the bits?

  11. Hmmmm. KF has basically rehashed the UPB, asserted a number and then not addressed the eleP(T|H)ant in the room.

    D-, Must revisit probability fundamentals.

  12. VJT:

    Thornton (and others) demonstrated that a protein sequence (let’s say a duplicated gene) can acquire a new binding function with as few as three point mutations. When you are calculating CSI, does the new protein have three bits of information, or more?

    If you do not know the history of the sequence, how can you tell how many bits were poofed into existence by the designer, and how many are the result of a chain of small changes?

  13. VJT:

    You are a philosopher, so I hope you appreciate the problem; that you are using an assumption about the history of a sequence to prove something about the history of the sequence — namely that it could not be the result of accumulated changes.

    I freely admit that evolutionary biologists also make assumptions. the difference is that biologists go into the laboratory and see if their assumptions can be supported by observation and experiment.

  14. Hi Petrushka,

    Back again. Just a few quick answers to your queries, as I’m rather busy now. Re Douglas Axe and citrate, you might like to read these articles:
    http://www.evolutionnews.org/2011/09/richard_lenskis_long_term_evol051051.html
    http://www.biologicinstitute.org/post/32246480851/innovation-or-renovation

    “But how significant was this innovation? In his paper in Quarterly Review of Biology, Dr. Michael Behe pointed out that E. coli was already capable of using citrate for anaerobic growth (when no oxygen was available). He postulated that a change in gene regulation could turn on citrate transport and permit growth on citrate under aerobic conditions.

    “After an enormous amount of work, having sequenced the genomes of many clones along the lineages that led to the ability to use citrate, as well as lineages that never did, and testing the phenotypes of identified mutations, Blount et al. have now reported that Behe was largely right. The key innovation was a shift in regulation of the citrate operon, caused by a rearrangement that brought it close to a new promoter….

    “The total number of mutations postulated for this adaptation is two or three, within the limits proposed for complex adaptations by Axe [2010] and Behe in Edge of Evolution. Because the enabling pre-adaptive mutations could not be identified, though, we don’t know whether this was one mutation, a simple step-wise series of adaptive mutations, or a complex adaptation requiring one or two pre-adaptations before the big event.

    “But does this adaptation constitute a genuine innovation? That depends on the definition of innovation you use. It certainly is an example of reusing existing information in a new context, thus producing a new niche for E coli in lab cultures. But if the definition of innovation is something genuinely new, such as a new transport molecule or a new enzyme, then no, this adaptation falls short as an innovation. And no one should be surprised.”

    See also:
    http://www.evolutionnews.org/2013/07/douglas_axe_pro074781.html
    http://www.evolutionnews.org/2013/07/a_response_to_m074821.html
    http://www.evolutionnews.org/2013/08/a_challenge_for075611.html

    Re origin of the Italian language: see http://linguistics.byu.edu/classes/ling450ch/reports/Italian2.html
    The only event which might qualify as a linguistic saltation is the invasion of the Ostrogoths in the fifth century.

    Re origin of Basque: nobody seems to know anything about its relatives, some of which are likely extinct. See http://en.wikipedia.org/wiki/Basque_language#Hypotheses_on_connections_with_other_languages

    Hope that helps.

  15. Hi Petrushka,

    The figure of 500 bits for a typical protein can be found in Stephen Meyer’s “Signature in the Cell.” From memory, it’s either chapter 9 or 13. If a protein undergoes “microevolution” and becomes a more efficient molecule, how do I count the bits? That’s fairly straightforward: from the number of changes required for the improved functionality.

    If a protein sequence can acquire a new binding function with as few as three point mutations, then I guess I’d say that’s three new bits (not very remarkable).

    Re your point about having to know the history of a protein in order to infer that it was designed: what we really need to know is the smallest possible size of the chasm separating it from its nearest functional neighbor. In cases where we can show that the chasm is great, a design inference is warranted. Given that the smallest functional protein has to contain 100-odd amino acids, and given that singleton proteins differ radically from their nearest neighbors, it seems that we have outcomes here which could only have been produced by design.

    That’s all for now.

  16. Does anyone see anything about primary, secondary, tertiary and higher level structures in any of these ID/creationist calculations? What about function and behavior? Where is the temperature dependence?

    Do we see anything in ID/creationist calculations about potentials describing the interactions between atoms in the system? What about spin-spin couplings that are measured from NMR experiments? What about the aqueous environments in which such molecules form? Do we see anything about free energies? Is there anything about x-ray diffraction studies of molecules? Is there anything about the classical and quantum mechanical regions of calculations?

    Compare ID/creationist calculations with, say, something like this.

    We see lots of papers cited by ID/creationists; but to what purpose? Can any ID/creationist explain the relevance of any of their “calculations?”

  17. vjtorley

    “If a protein sequence can acquire a new binding function with as few as three point mutations, then I guess I’d say that’s three new bits (not very remarkable).”

    QED? Evolution. Do you have an example on something “non-remarkable” emerging?

    “Given that the smallest functional protein has to contain 100-odd amino acids”

    False. 4-Oxalocrotonate tautomerase, an enzyme, is 62 amino acids. Insulin is 51 amino acids. The ankyrin repeat motif, found in ~5% of proteins is 33 amino acids. Trp cage, from the saliva of Gila monsters is 20 amino acids. A 1 amino acid “protein”: proline is catalytic. Large symmetrical proteins have been deconstructed into short peptides (ever wonder why so many proteins are repeat proteins (of short simple motifs or have clear internal symmetry).

    “what we really need to know is the smallest possible size of the chasm separating it from its nearest functional neighbor. ”

    Indeed. And for that, we need close neighboring genome sequences. Comparing the 1 crustacean and 1 amphibian genome will yield a lot of lonely genes. One species of Drosophila and another, not so much. And I think those that remain are quite explicable.

    Why don’t one of you design types point out a few genes in closely related species that MUST be designed? Disprove macroevolution. Let me know….

  18. vjtorley: Re origin of the Italian language: see http://linguistics.byu.edu/classes/ling450ch/reports/Italian2.html
    The only event which might qualify as a linguistic saltation is the invasion of the Ostrogoths in the fifth century.

    I’m not sure what point Petruska is heading towards, as I haven’t read the relevant comments over at UD, but the best language to look at for sudden change is probably English. For example, close to 30% of English words are of French origin. 10,000 words alone were inherited during the Norman invasion.

    There is also a phenomenon known as “the great vowel shift” which occurred in a relatively quick time-frame, 300 years.

  19. vjtorley, to petrushka:

    Re your point about having to know the history of a protein in order to infer that it was designed: what we really need to know is the smallest possible size of the chasm separating it from its nearest functional neighbor. In cases where we can show that the chasm is great, a design inference is warranted. Given that the smallest functional protein has to contain 100-odd amino acids, and given that singleton proteins differ radically from their nearest neighbors, it seems that we have outcomes here which could only have been produced by design.

    Vincent,

    To show that “the chasm is great” between a protein and its nearest functional neighbor, you have to know where the nearest functional neighbor is.

    We haven’t exhaustively explored protein space, so any claim regarding a “nearest functional neighbor” is highly suspect.

    You also seem to be confusing the “nearest functional neighbor” with the “nearest functional neighbor in an extant organism”. The latter is irrelevant to the size of the “chasm” that evolution must have crossed. What matters is the size of the chasm between a protein and its ancestral forms, and those ancestral forms are typically unavailable for inspection due to extinction.

    Also, how do you determine whether a nearby protein is “functional”? What function, and how functional? A protein that is non-functional in one context may be functional in another, and the function may change over time. Marginally functional proteins may be sufficient in some contexts and insufficient in others.

  20. davehooke:

    Obvioulsy you are not familiar with the CSI calculations of cake and aardvarks by Mr J Gallien.

    No; I am not.

    Apparently the Nobel Committee isn’t familiar with them either.

  21. KF makes a specific claim that FSCO/I in paragraphs of text can be quantified.

    As this search will show:

    site:www.uncommondescent.com “billions of examples”

    It would seem to be to be simpler to generate a specific figure for FSCO/I for a comment rather then an organism. That method can then be extended to proteins if shown to actually work.

    KF, will you demonstrate the calculation of FSCO/I for a number of text strings? Pick any 10 comments at UD and calculate the FSCO/I for them. Then do the same for 10 comments here.

    Or pick any other 10 out of the “billions of examples” you claim exist.

    6 –> In short, the only empirically and analytically known — on billions of examples and zero credible counter examples — source of FSCO/I is intelligence. So, as with the laws of thermodynamics, we are entitled to infer on that pattern until and unless a credible counterexample is presented.

    For example.

    I can’t present you with a counterexample because I cannot measure the FSCO/I being output by any process I might present, and you seem unwilling to demonstrate this (yet it can apparently be done for billions of example!).

    Seems we need a way to measure FSCO/I. If it is present, the source of that data must have been intelligence, right KF?

    Yet if you insist on knowing about the source of the data in advance (this long comment written in English seems to make sense and be in English – it must have been the product of intelligence!) then what validity do your claims have?

    You need a metric that can be applied to arbitrary data and that shows the level of FSCO/I present. Or are you telling me with billions of examples out there you’ve yet to calculate the FSCO/I for a single one of those billions of examples.

    As it seems to me that of your billions of examples knowing the language they are written in would be a help when deciding if they are the products of intelligence or not. And if your claim is you can only recognise design when you recognise design you are already familiar with, then that’s a trivial claim.

    Or if you are not up to that challenge, KF, simply demonstrate that two strings of equal length can have different values for FSCO/I and that will be a start….

  22. vjtorley,

    Re origin of Basque: nobody seems to know anything about its relatives, some of which are likely extinct.

    So when a language stands isolated, you are happy to assume common descent with stepwise, ‘undirected’ change and extinction of intermediates. When a protein does, it’s Designer Intervention.

    I’m sure you can see why this might be seen as arbitrarily inconsistent.

  23. vjtorley,

    If you assess a protein based upon its bitwise composition, you will get the same value for every protein of a given size, regardless of function. You might also over-inflate, if your protein actually consists of submodules. Which ALL functional proteins do.

    Proteins (many, not most) are long because selection tends to favour greater length – it allows for more subtle binding-release kinetics, and more sites for cofactors. But less-subtle kinetics and fewer sites in shorter-chain ancestors does not mean they could not function, nor extend to the modern bulk. It’s like saying nothing, short of a 747, can fly. (Let’s not over-extend that metaphor, eh? 😉 )

  24. VJT:

    Unless you can demonstrate otherwise, Axe could labor with his methods for a thousand years and not find the stepping stones to citrate metabolism found by evolution in the Lensky experiment.

    Trivializing the result does not help the design hypothesis, because it implies that designers can’t even find the simplest bridge to new functionality. Evolution is demonstrably smarter than Douglas Axe, even when the design task is trivial.

  25. Language evolution is interesting because it is well documented. It happens not because anyone tries to make it happen, and it seems to happen despite efforts of grammarians to prevent it from happening.

    The changes in spelling are particularly interesting, because spelling is analogoius to the genetic code.

    Spelling has changed without preventing people from reading. Typos like “pwned” have entered the language because they add function. In that particular case, the added function is humor. No designer is in charge of which typos get selected for common usage. The selecting hand is invisible, and it often overrides attempts to purify the language.

  26. Allan Miller:
    vjtorley,
    So when a language stands isolated, you are happy to assume common descent with stepwise, ‘undirected’ change and extinction of intermediates. When a protein does, it’s Designer Intervention.
    I’m sure you can see why this might be seen as arbitrarily inconsistent.

    I’m glad to see that someone understands the point. Isolation from cousins does not imply poof. We have numerous detailed histories of language evolution, and one or two languages whose ancestry is baffling. I have asked about this repeatedly, particularly to gpuccio. His argument rest entirely on the apparent isolation of protein domains, and he has just in the last couple of days, affirmed his belief that protein domain sequences were poofed into existence.

    This touches on the one aspect of my argument that is non-negotiable. If you believe in poof, you will not design experiments to test whether non-poof is possible. You will not try, as Thornton et al have done, to reconstruct ancestral histories. You will not test whether stepping stones can explain divergent histories.

    Basically, you will not do science. This is why ID is a science killer. It is why, evolution is the only explanation that can be investigated.

  27. petrushka:
    Language evolution is interesting because it is well documented. It happens not because anyone tries to make it happen, and it seems to happen despite efforts of grammarians to prevent it from happening.
    . . .
    No designer is in charge of which typos get selected for common usage. The selecting hand is invisible, and it often overrides attempts to purify the language.

    I suspect the Académie française think of themselves as purifying selectors as well. 😉

  28. VJT:

    The total number of mutations postulated for this adaptation is two or three, within the limits proposed for complex adaptations by Axe [2010] and Behe in Edge of Evolution. Because the enabling pre-adaptive mutations could not be identified, though, we don’t know whether this was one mutation, a simple step-wise series of adaptive mutations, or a complex adaptation requiring one or two pre-adaptations before the big event.

    The key element of this discussion is the one or two enabling mutations. Unless Axe can bridge this gap, his work is not relevant to supporting or undermining evolution. It’s really that simple. I could write a six or ten thousand word essay expanding on that, but the point is made. Axe’s work is simply not probative.

    Even in the eye-blink of ten years in a few test tubes, evolution chained at least three mutations — including at least two neutral ones — to produce an effect. Neutral duplication, enabling neutral mutation, beneficial mutation. Unless Axe actually tests plausible evolutionary scenarios, his work has no value to science.

    Within Behe’s Edge, but in only ten years, in a tiny population. New protein domain sequences appear less often that once in a million years.

  29. Joe@UD,
    Link.

    And this should answer OMagain’s question:

    So you are saying that you can measure the FSCO/I in a protein but can’t measure it in a paragraph of text?

    Also, I find it odd that “FSCO/I” does not appear in the article you link to, neither does “CSI”. Why not?

    So if you can indeed measure FSCO/I did it increase or decrease in Lenski’s experiment after the citrate mutation occurred? Please show your working!

  30. Patrick: I suspect the Académie française think of themselves as purifying selectors as well.

    I had them in mind. It takes a lot of designers to block evolution. While you have your finger in the dike, the ocean is seeping in under your feet.

  31. Joe@UD

    All that happened was an existing gene that encoded for a protein that aids in getting the citrate into the cell when O2 is not present, ie an anaerobic environement, was duplicated and put under control of a promoter that was active in the prsence of O2.

    Ah, is that all?

    When that gene was duplicated and put under the control of a promoter, did the FSCO/I go up or down?

  32. OMagain:
    Joe@UD,
    Link.

    So you are saying that you can measure the FSCO/I in a protein but can’t measure it in a paragraph of text?

    Also, I find it odd that “FSCO/I” does not appear in the article you link to, neither does “CSI”.Why not?

    So if you can indeed measure FSCO/I did it increase or decrease in Lenski’s experiment after the citrate mutation occurred? Please show your working!

    This is giving me flashbacks.

    If they follow the usual pattern, and intelligent design creationists are nothing if not traditionalists, they will continue to claim to be able to calculate CSI or one of the alphabet soup of variants for another few days. They will then claim to have done so. When asked to point to where this act of creation (See what I did there?) took place, they will either repeat the claim, without reference, or resort to insults (or both).

    The odds of getting an actual calculation from kairosfocus exceed Dembski’s universal probability bound.

  33. gpuccio:

    The opposite is not true. Neutral mutations cannot help in traversing the space between islands of functionality, for two important reasons:

    a) They can go in any possible direction, because they are random. Even considering the allelic effect of genetic drift, the result remains completely random. Therefore, neutral mutations in no way help in overcoming the probabilistic barriers: the number of possible states to be tested remains the same.

    b) Negative NS definitely acts against the traversing. When you argue about “lose money and argue that because he lost, it is impossible to win”, you are forgetting the exacting role of negative NS.

    The number of states being tested is always limited to those in the immediate vicinity. That is not an astronomical number.

    Once a duplication occurs — as in the Lensky experiment — negative selection is pretty much irrelevant. Most mutations in the duplicate are going to be neutral. Again, Lensky.

    Until Axe accounts for the combined possibilities of duplication and drift, his work is not probative.

  34. gpuccio:

    Regarding your “Experimental Rugged Fitness Landscape in Protein
    Sequence Space”

    You are the proponent of intelligent selection. You have argued it is more powerful than natural selection, and I have argued the opposite. You present evidence that goal directed design is hard.

    Your conceptual problem is assuming there is a goal. You ignore sideways meandering.

  35. gpuccio:

    6) Regarding languages, I don’t think I understand your point. Languages are structures which are formed and evolve in conscious intelligent beings. Why do you use them as models for non designed evolution?

    Because languages evolve. They are not designed.

    Designed languages like Esperanto do not survive in the wild.

    I do not know for certain your native language, but you are familiar with English. Despite centuries of effort by designers, no one is in charge of English spelling and grammar. Efforts to rationalize and standardize language fail over time. Usage and spelling change.

    ID proponents like the analogy with computer code and like to point out that a single bit error in a computer program will crash the system.

    But language is a much better analogy. Errors occur all the time and do not crash the system. Some of them get incorporated into the language. Over time, languages change to the extent that speakers of a language can’t read or speak the earlier version.

    Now about Basque. Unless you believe that the Basque language was poofed into existence by a designer, it evolved by incremental change.

    But it has no apparent continuity with any other extant language. It has no living cousins. It is a demonstration that a system of coding sequences can change incrementally over time, to the point that cousinship cannot be determined.

  36. keiths:

    davehooke:

    Let’s share with everyone:

    aardvark

    cake

    obvioulsy

    Ah, that Joe G; the one who was banned from this site because of his porno links.

    I haven’t kept up with his antics; but I think he can forget about a Nobel.

  37. Patrick: This is giving me flashbacks.

    Interestingly, there is another “tell” in their calculations that you provide with that link.

    If you look at their examples of “genes,” you will note that they don’t divide by all the factorials of the numbers of repetitions of all identical letters. After all, if one interchanges two identical letters, it is the same string; the n! permutations of the n repetitions of a given letter produce the same string. That goes for every letter that appears twice or more.

    And what is the binding energy of A next to T as opposed to A next to G, etc.? In other words, why to they think the probabilities are always the reciprocal of the number of possible letters per site?

    So even if any of these calculations were relevant to something, they are counting incorrectly and ignoring binding energies; and that’s just the least of their problems.

  38. I think more importantly, they do not know how many characters are significant and how many changes could be tolerated. This gets a bit ridiculous if the main mode of invention is duplication followed by mutation of the dupe.

    In which case the only relevant number is the number of changes since the duplication.

    They also ignore the inconvenient fact that most of the heavy lifting in terms of gene invention has been done by bacteria. That means the process has involved large populations.

  39. Friday nights are generally rather slow, and I may not be available to post much for the next 24 to 36 hours. So I’m going to summarize where I think the discussion has been.

    The most convincing argument I see is against any straight path from minimal functionality to optimal functionality. Call it the problem of local maxima.

    To this I would respond that this is an argument against design. In particular it is an argument against gpuccio’s concept of directed or intelligent selection. I would argue that straight line paths up Mount Improbable are very improbable.

    This makes goal directed design very hard and very unlikely.

    But it doesn’t impede evolution, because evolution is not goal directed. Evolution wanders sideways more than upward. It does not “need” to leap out of local maxima, because populations are always alive and functional, or they would be extinct. When something in the environment changes and makes adaptation necessary, the most likely result is extinction. Intelligent selection is not likely to come to the rescue.

    The Lensky experiment directly addresses this problem and illustrates how evolution can work sideways and sometimes find a path around a barrier.

    But it is important to note that the populations in Lensky’s experiment were not goal directed and did not need to change. Nor was there any way for an intelligent selector/designer to know which neutral mutation would be the enabler.

    This is the fatal flaw in Axe’s approach. He does not consider sideways evolution.

  40. Petrushka,

    Is it just me, or are informed discussions about what Axe’s work actually does and proves dead-on-arrival (especially the dreadful work done with Gauger)?
    Some prolific posters just seem to vanish when the discussion gets good. This, despite that Axe’s work “disproves” molecular evolution is one of the DI’s most repeated points: e.g. Luskin, just today: http://www.evolutionnews.org/2014/03/cosmos_episode_083331.html

    Perhaps a collaborative post some day. I’ve posted this outline before:

    I. What evolution is:
    a. Not goal directed (no given change is preordained or required to happen)
    b. Not the interconversion of one modern enzyme into another modern enzyme
    Creationist Todd Wood:
    “Gauger and Axe focused directly on converting an existing enzyme into another existing enzyme. That left me scratching my head, since no evolutionary biologist would propose that an extant enzyme evolved directly into another extant enzyme. So they’re testing a model that no one would take seriously? Hmmm…”
    http://toddcwood.blogspot.com/2011/04/protein-evolution-in-bio-complexity.html
    c. An alternative approach yields success: Thornton et al. trace evolution by ancestral reconstructions:
    http://www.ncbi.nlm.nih.gov/pubmed/17702911
    http://www.ncbi.nlm.nih.gov/pubmed/23798447
    “shift in function was driven primarily by two historical amino acid change”
    This approach provides a molecular basis for why Axe’s work fails:
    ” Unless these ratchet-like epistatic substitutions are restored to their ancestral states, reversing the key function-switching mutations yields a non-functional protein.”
    http://www.ncbi.nlm.nih.gov/pubmed/19779450

    II. The big number game: 10^whaa? and molecular evidences of a smaller era
    a. The abundance of repeat proteins: (~25-50 amino acid motifs) in modern proteins
    b. The symmetry of smaller motifs in modern proteins
    Experimental Deconstruction:
    “complex protein architecture was an early evolutionary achievement
    involving oligomerization of smaller polypeptides.”
    http://www.mikeblaber.org/101007-s00018-012-1077-3.pdf
    c. Enzymes from simple folds: a hexamer of a 62 amino acid protein, 4-Oxalocrotonate tautomerase, is catalytic. Relatives in the same fold family have different activities. Sequence requirements appear limited.

    III. Critiques of Axe
    a. Cherry picking: his own data
    “These results imply that hydrophobicity is nearly a sufficient criterion for the construction of a functional core and, in conjunction with previous studies, that refinement of a crudely functional core entails more stringent sequence constraints than does the initial attainment of crude core function. Since attainment of crude function is the critical initial step in evolutionary innovation, the relatively scant requirements contributed by the hydrophobic core would greatly reduce the initial hurdle on the evolutionary pathway to novel enzymes.”
    http://www.ncbi.nlm.nih.gov/pubmed/8643620
    b. Cherry picking: ingoring other estimates based on experimental data
    -Vary from 1 in 10^10 and less.
    c. Methodologies
    i. Designed to fail part 1: a barely functional enzyme gets beat on some more
    ii. Designed to fail part 2: Interconversion of 2 extant enzymes with only !!34%!! identity by only a handful of point mutations is destined to fail.
    iii. Gauger’s tryptophan experiment: A double mutant of a Trp-producing enzyme was overexpressed in cultures where the key amino acid, tryptophan was provided at a sufficient concentration for growth and survival. Protein overexpression is a burden on cultures. Therefore, mutations that block the expression of the highly overexpressed, inactive Trp-producing enzyme accumulate, with the effect of doubling the growth rate in 5 generations. Conclusion: Trp synthesis didn’t re-evolve. (in the absence of selective pressure for Trp synthesis). But did the culture evolve? A doubling of fitness in 5 generations–please ignore that, says Gauger.

  41. The really big problem is that IDists only consider goal directed evolution.

    Improvemments and such.

    Because of this they do not realistically project histories.

Leave a Reply