Request for Criticisms: A Tutorial on Specified Complexity

I am working on a series of tutorials to cover the basics of Intelligent Design, especially the mathematics of it. This is my tutorial on Specified Complexity, and I would appreciate any thoughtful criticism of it.

Note that I am specifically requesting criticisms on the content of the video itself, not on applications of the concept that are outside the bounds of the video. This is both to help me (I’m trying to improve my presentation of ID) and to help clarify the conversation (is the criticism of *this* information or of some *other* information).

388 thoughts on “Request for Criticisms: A Tutorial on Specified Complexity

  1. Acartia: Design can never be excluded.

    If you have a way to exclude design that doesn’t beg the question, I am all ears.

    What objective empirical evidence do you have that snowflakes are not the product of design? If any. If you can’t provide any then you need to retract your claim. Just ask Patrick.

  2. “Mung: Haha. Funny. You see the word SELECTION and don’t notice the AGENCY lurking behind it.”

    I wonder what’s the psychological purpose of playing these word games when deep down you know you’ve lost?

  3. fifthmonarchyman:
    Hey Johnnyb,

    What is the best way to contact you?

    I know of a bright undergrad who is double majoring in theoretical math and computer science that would be very interested in an internship.

    Intelligent design creationism destroying Winston Ewert’s career before it started isn’t enough? You want to sacrifice another student on that altar?

  4. colewd:
    . . .
    Lenski did 30 years of experiments with bacteria for the equivalent of 1.5 million human years of evolution.In the end the bacteria remained bacteria.

    Your mask is slipping. The creationist underneath is peeking out.

  5. Mung:

    Videos strike me as similar to the debates creationists love to challenge scientists to have. The goal is persuasion and point scoring, not understanding and constructive discussion.

    Wait. Don’t tell me, let me guess. You haven’t watched the video. Am I right?

    I made it very clear that I did not and will not, and why, in the section you failed to quote.

    Your point scoring exercise failed.

  6. colewd,

    Allan, the flagellar is 100k of DNA sequential space that needs to become organized to produce proteins that spin a motor 100k rpm. 30 proteins need to be assembled and form together in shape and charge and all work together for a single purpose. The odds of this happening by anything but purposeful design is so remote that counting homologies is like the Trump re count in Pennsylvania.

    If there is any homology at all, then this is a damned curious fact, on a design view. The Designer has the entirety of protein space to go at (with, to repeat, no obvious means by which ‘intelligence’ tells it anything about where to find it) and yet it repeatedly mined the same corner for multiple components. This protein space that has hardly anything in it, but does have a whole bunch of neighbours just here – just where the Designer found ’em.

    If, OTOH, the apparent homology indicates real common descent at the protein level, your invocation of BigNumbers as if the entirety were assembled by plugging random monomers together dipped from a sack would be rather irrelevant. If you want to critique the natural process, it’s no good strawmanning it. You have to take the case on terms that make sense within the paradigm.

  7. colewd: The odds of this happening by anything but purposeful design is so remote

    And what can we infer about such a designer?

  8. Mung: What objective empirical evidence do you have that snowflakes are not the product of design?

    According to my calculations, snowflakes are not superspecificandcomplexialidocious enough to be designed

  9. Mung: How did you determine scientifically, mathematically, however … that a snowflake is not designed?

    Dembski says snowflakes are not designed. That’s good enough for me. 😉

  10. colewd,

    Here is a reference paper on the protein make up of the motor.

    It also discusses evolutionary history, extensive homology and the surprising (on a Design paradigm) congruence of the flagellar protein tree upon a species tree constructed from non-flagellar proteins. I’d say it supports my contentons much more than yours.

  11. colewd:
    Allan Miller,
    Create a test that validates the node and I’m all in.The rest is speculation.

    It isn’t clear what you mean by “validates”, but the typical tests for such situations are bootstrap and likelihood ratio tests. Why are they not sufficient?

  12. John Harshman: It isn’t clear what you mean by “validates”, but the typical tests for such situations are bootstrap and likelihood ratio tests. Why are they not sufficient?

    You know full well that if the tests did not show the high probabilities of relatedness that they do, the IDists would crow that this proves that evolution didn’t happen.

    For once, they’d be right.

    But get what evolutionary theory predicts, and the excuses abound.

    Glen Davidson

  13. Mung: How did you determine scientifically, mathematically, however … that a snowflake is not designed?

    He probably used Boolean probability, factoring in the physics of ice crystal formation and the randomness of snowflake nucleus formation and tentatively concluded that invoking a designer wouldn’t add any explanatory clarification.

    But nobody can be sure that the Abrahamic god didn’t design every damn snowflake that ever formed and will continue to do so until our sun dies. When you’re infinite, like the Big Guy, you can spend an eternity on that kind of thing.

    And why deny the Big Guy credit? He might not like you doing that.

  14. A few notes from the discussion, since I’ve been out of it for a bit.

    I should point out that, while I do disagree with the grand unified tree hypothesis, there is literally nothing in Intelligent Design on its own which precludes the possibility of a universal common ancestor. Now, there might be practical constraints (i.e., the size of genome that would be required to have a universal common ancestor may be way too big), but design theory itself does not say specifics. So, an argument against ID saying “it looks like its ancestors” is not actually an argument against ID. It might be an argument *for* common ancestry, but not against ID.

    I think people read *way* too much into ID. It is a precise tool for a precise task. It’s about identifying, quantifying, and analyzing design-as-design. The idea that it is a full replacement theory is simply untrue. It is, at most, a component of such a theory.

    The reason I hate the conflagration of ID with creationism is not because I think I’ll be labelled a fundy (I *am* a fundy and rather proud of it), but rather I think it muddies the intellectual waters, so that people don’t even know what they are talking about. When people use complaints against creationism as if they are complaints against ID, it is just so much noise. When people use the word Intelligent Design when they mean creationism (including by creationists) it again muddies the waters making noise.

    What makes ID interesting is that it *doesn’t* aim to be a full view of life, but rather attempts to restrict itself to one aspect of the question.

  15. johnnyb:

    What makes ID interesting is that it *doesn’t* aim to be a full view of life, but rather attempts to restrict itself to one aspect of the question.

    That’s also what makes it worthless as a scientific concept with zero explanatory power.

    I’m still curious to see how your specified complexity calculations fare when applied to iterative processes involving feedback.

  16. “The reason I hate the conflagration of ID with creationism”

    is because you don’t understand that ID was a renaming of creationism as a legal strategy, which failed in 2005, which is why ID has been withering for 11 years?

    The authors of Of Pandas and People certainly understood that ID and creationism were different names for the same thing.

  17. I can understand people wanting ID to be science. But if you look at the history of science, and how it works, and the history of ID, you should be able to see that ID is a pseudoscience scam. If you can’t, well, there’s still hope for you, but I don’t expect much.

  18. Dembski’s fake “science journal” PCID stopped publishing in December 2005. The month the Dover decision came down. Does that sound like an actual scientific journal to anybody? When the legal strategy failed, they lost a lot of effort.

  19. johnnyb: It is a precise tool for a precise task. It’s about identifying, quantifying, and analyzing design-as-design

    Do you have an example where this was actually done then?

  20. colewd: Lenski did 30 years of experiments with bacteria for the equivalent of 1.5 million human years of evolution. In the end the bacteria remained bacteria.

    This is up there among the dumbest statements ever uttered on this site, which is saying something.

    Bacteria is a DOMAIN of life, Human is a GENUS. Reversing the roles, you’re basically saying Lenski’s experiment should have resulted in Homo Sapiens becoming something else than EUKARYOTES in 1.5 million years. Why in the flying fuck would you expect or demand that?

    Yes, the bacteria in Lenski’s experiment still belong to the DOMAIN bacteria. Just like Homo Sapiens still belong to the DOMAIN eukaryota, despite the DOMAIN having existed for at least 2 billion years. Are we now to think there has been little to no evolution within the eukaryotic domain, just because they’re still eukaryotes?

    The degree of difference, both genetic and phenotypic, between the species of E coli bacteria Richard Lenski started his experiment with, and the bacteria that are growing in his flasks now, is AT LEAST as great as the difference between Homo Sapiens now, and our ancestor 1.5 million years removed.

  21. johnnyb: I think people read *way* too much into ID. It is a precise tool for a precise task. It’s about identifying, quantifying, and analyzing design-as-design. The idea that it is a full replacement theory is simply untrue. It is, at most, a component of such a theory.

    This is confusing. So ID is a tool, not a theory, that might be a component of some other theory. What theory is that?

  22. dazz: This is confusing. So ID is a tool, not a theory, that might be a component of some other theory. What theory is that?

    Instead of ID, I suspect he meant to say specified complexity.

  23. johnnyb: I think people read *way* too much into ID. It is a precise tool for a precise task.

    Which is why it’s weird that in your tutorial you give up on the outset any specific application to it. Without application, whatever it is you’re explaining is useless, and your tutorial about it is also useless. Redo the whole thing, show how it’s a precise tool for a precise task and exactly what task it is for.

    Show the thing in live action. It would be the best way from the pedagogical point of view.

  24. johnnyb: I think people read *way* too much into ID. It is a precise tool for a precise task.

    A “precise tool” for saying “it’s designed” whenever complex functionality is found. It really does nothing but conflate functional complexity and design, while ignoring the fact that life is missing the ordinary traits (especially those coming from being able to think beyond merely tweaking inherited information) that lead us to conclude that intelligence was involved in its making. In the honest sense of “complexity”–rather than using Dembski’s redefinition of “complex” to mean “unlikely to be produced without intelligence”–ID would be incapable of identifying normal tools and other artifacts in archaeology as having been manufactured by humans, because these are almost invariably rather simple.

    It’s about identifying, quantifying, and analyzing design-as-design.

    No, or it could actually identify a simple knife as designed. It can’t do so using an honest definition of “complex,” and it has no purpose other than to claim that life was designed even though it looks like it evolved without anything like intelligence transcending and overriding evolutionary processes.

    The idea that it is a full replacement theory is simply untrue.

    Then why is it sold as an alternative to evolutionary theory? Of course it’s meaningless tripe unworthy of the name “theory,” but Stephen Meyer wrote two pages (pp. 391-392) in Darwin’s Doubt on “Reasons to regard Intelligent Design as a Scientific Theory,” using that subtitle to introduce his discussion.

    It is, at most, a component of such a theory.

    So would the full theory be called “creationism”? And why doesn’t Stephen Meyer know what you know?

    Glen Davidson

  25. johnnyb: I think people read *way* too much into ID. It is a precise tool for a precise task. It’s about identifying, quantifying, and analyzing design-as-design.

    Johnnyb has just given the secret password that will allow anyone to pass the ID Turing test.

    If you understand ID you understand this simple statement. Those on both sides who don’t understand it will inevitably slip up and portray it as more than this.

    Peace

  26. GlenDavidson: ID would be incapable of identifying normal tools and other artifacts in archaeology as having been manufactured by humans, because these are almost invariably rather simple.

    I would agree that ID is incapable of this as it is currently understood. The solution is not to abandon the exercise and throw up our hands it’s to modify it look for other ways of identifying design.

    peace

  27. fifthmonarchyman: I would agree that ID is incapable of this as it is currently understood. The solution is not to abandon the exercise and throw up our hands it’s to modify it look for other ways of identifying design.

    What is design? Next: What makes you say it can be identified? Identified as in distinguished from something else? From what else?

  28. fifthmonarchyman: I would agree that ID is incapable of this as it is currently understood. The solution is not to abandon the exercise and throw up our hands it’s to modify it look for other ways of identifying design.

    peace

    We do archaeology (you really don’t read very thoroughly, do you?), meaning that we already know how to identify design far better than does ID. And I don’t mean “we” in the sense of including you.

    But real indications of design, such as using particular designs to fit particular needs, rather than tweaking vertically inherited form and function to fit particular needs (as is typical in vertebrate evolution), are missing in life. IDists don’t like that.

    Glen Davidson

  29. GlenDavidson: Then why is it sold as an alternative to evolutionary theory?

    Because those who are trying to “sell it” as an alternative don’t understand it.

    GlenDavidson: Of course it’s meaningless tripe unworthy of the name “theory,” but Stephen Meyer wrote two pages (pp. 391-392) in Darwin’s Doubt on “Reasons to regard Intelligent Design as a Scientific Theory,” using that subtitle to introduce his discussion.

    I have a theory as to why this is the case. 😉

    Seriously why the resistance to calling ID a theory? It’s a theory according to the dictionary definition

    Theory- The analysis of a set of facts in their relation to one another.

    Calling ID a theory does not mean it is a replacement for Evolutionary Theory any more than speaking of the Gene Theory means you reject Darwinism.

    http://biology.about.com/od/geneticsglossary/g/genetheory.htm

    peace

  30. GlenDavidson: We do archaeology (you really don’t read very thoroughly, do you?), meaning that we already know how to identify design far better than does ID.

    Archaeology is one place where we attempt to identity design there are others as well like forensics.

    If ID could generalize the inference it would be a good thing right?

    GlenDavidson: But real indications of design, such as using particular designs to fit particular needs, rather than tweaking vertically inherited form and function to fit particular needs

    So are you saying that it is impossible to design in this way? Or are you only saying that there is no possible way to identify design that is accomplished in this fashion?

    peace

  31. Erik: What is design? Next: What makes you say it can be identified? Identified as in distinguished from something else? From what else?

    I think that what we are really talking about is inferring consciousness in the source of an object or event. That is why I’m so interested in Turing tests right now.

    ID is an attempt in my opinion at developing a Turing test for the author/source of life and the universe.

    peace

  32. fifthmonarchyman: I think that what we are really talking about is inferring consciousness in the source of an object or event. That is why I’m so interested in Turing tests right now.

    ID is an attempt in my opinion at developing a Turing test for the author/source of life and the universe.

    Amazing, it surely has evolved a bit since it meant calculating a complexity or detecting something. One can hardly recognize it. One might even ask if it’s the same thing.

  33. Erik: Amazing, it surely has evolved a bit since it meant calculating a complexity or detecting something.

    Not really calculating a complexity is simply one proposed method for conducting a Turing test it’s not the only one. And you are detecting something that something is consciousness in my opinion.

    peace

  34. dazz: ID is whatever the fuck you want. Priceless

    Instead of spewing profanity why not explain why ID is not a Turing test?

    peace

  35. fifthmonarchyman: Instead of spewing profanity why not explain why ID is not a Turing test?

    peace

    I already said why it’s all pointless, and you just confirmed what I said: namely, that ID is so devoid of an explanatory narrative that it’s whatever the fuck you want it to be… in your opinion, a Turing test..

    Hey! Look how I pass your test:

    “ID is an attempt at developing a Turing test for the author/source of life and the universe”

    yippee!

  36. Looks like we’re all agreed ID / specified complexity is absolutely worthless when it comes to detecting design in any real world inanimate object, let alone detecting “design” in biological life. But if we ever run into a gang of crooks trying to rig the results of a coin flipping contest we’ll nail their asses cold. 🙂

  37. johnnyb:
    I think people read *way* too much into ID. It is a precise tool for a precise task. It’s about identifying, quantifying, and analyzing design-as-design. The idea that it is a full replacement theory is simply untrue. It is, at most, a component of such a theory.

    ID is neither precise nor a tool. It’s only task is to avoid the separation of church and state in the U.S.

    If you disagree, please take up the challenges to demonstrate how to calculate CSI that have been posted in this thread.

    The reason I hate the conflagration of ID with creationism is not because I think I’ll be labelled a fundy (I *am* a fundy and rather proud of it), but rather I think it muddies the intellectual waters, so that people don’t even know what they are talking about.

    cdesign proponentsists

    When you can replace every instance of “creation science” with “intelligent design” in an entire book without changing the meaning, it is painfully clear that the two terms are synonyms.

  38. fifthmonarchyman: Archaeology is one place where we attempt to identity design there are others as well like forensics.

    Those are instances of human design. We know the capabilities and constraints of humans. What are the constraints of your god designer? If an entity can do literally anything then there is no way to determine if a particular artifact is the work of that entity.

  39. fifthmonarchyman: Instead of spewing profanity why not explain why ID is not a Turing test?

    Why don’t you explain why you think ID is a Turing test?

    Do you even know what a Turing test is?

  40. johnnyb,

    I plan to write a few more comments on your video. Thanks for posting. Regarding the coin flips, the improbability can be calculated based on the binomial distribution. We don’t have to have 100% heads, it can be some fraction, and the binomial distribution can deal with NON-equiprobable distributions as well.

    The homochirality argument fits well with the coin flip analogy for Specified Complexity in the OOL question. Less well known are situations were certain classes of amino acids must be located on a protein in such a way that when the protein is properly folded, an over abundance will occur in certain 3D locations.

    Below is conceptual depiction of 5 different kinds of transmembrane proteins that are shown as integral parts of the cell membrane.

    Notice the over abundance of “blue” amino acids (charged residues) on the outer parts of the cell membrane and scarcity of “blue” amino acids within the cell membrane.

    One can plug in the numbers of blue vs non-blue to some sort of binomial distribution, and I suspect the remoteness of the configuration would approach astronomical. I haven’t done the calculations. It would be worth doing as it shows the improbability of forming a transmembrane protein randomly with less of the difficulties of Doug Axe’s work.

    BTW, Bill Dembski occasionally uses the term Specified Improbability, which I think is a more accurate term than Specified Complexity. Although I’ve lobbied to drop all these unique terms from ID altogether in favor of more traditional terms in statistics.

  41. johnnyb,

    Now, there might be practical constraints (i.e., the size of genome that would be required to have a universal common ancestor may be way too big)

    Whaaat? It would not need to have anything more than the genes required for its immediate survival and reproduction.

  42. Patrick: Those are instances of human design.We know the capabilities and constraints of humans.

    Not necessarily. At the moment we don’t know how to differntiate between actions based on “will” and actions based on “predefined law”. Design requires “will”. We might be design tools for other beings, like Jesus.

  43. stcordova,

    I haven’t done the calculations. It would be worth doing as it shows the improbability of forming a transmembrane protein randomly with less of the difficulties of Doug Axe’s work.

    As usual, imagining the probability of something arising in one go.

    Alpha helixes have an amphipathic moment – an asymmetry in the distribution of hydrophobic and hydrophilic residues. This arises quite at random. If weak membrane binding occurs as a consequence and is favourable, there is clearly a selection pressure in favour of retaining further hydrophilic substitutions on one side, and hydrophobic on the other, which will enhance the binding. Thus an initially random perturbation is reinforced. Such helixes can form a pore, by forming bundles with the hydrophilic sides to the inner part – an orientation they will adopt very readily. And, amphipathic sequence can migrate from one protein to another, providing membrane binding for nothing, once it has arisen initially.

    It’s not an amino acid bingo machine.

  44. “I’m gonna make a 45 minute video tutorial about some fake science that doesn’t work!” is not the work of a smart person.

  45. I suspect Sal is just terrified of the idea that god doesn’t exist. He should get some courage.

Leave a Reply