The Intelligent Design Challenge – Dénouement

Ok, yeah, but you didn't address my detection of DESIGN in a genome, sir!

http://genomicron.blogspot.com/2008/02/evidence-of-design-in-genome.html

Ha! I raise you your DESIGN with VENTER (use TBLAST through the entire nucleotide database, you have to turn off complexity filtering, set word size to 2, expect threshold to 30000 and use the PFAM 30 database though]). Is it coincidence that almost everywhere you see DESIGN you also see VENTER. It is obvious that it was Craig Venter who designed the Chimp, and the dog and the mouse (and he is also to blame for some very nasty parasites too). Ordinary Design wins again (as to how Venter did it, possibly the Tralfamadorians took him back in time with a gene splice kit).

I thought that Dembski's filter determined that ALL life is designed. The watermark only shows that some sequences are doubly designed. This foregone conclusion is derived from the literal interpretation of a source Dembski need not identify, since it doesn't matter.

Well now, all we have to do is find a sequence that translates "GODDIDIT" in every organism and that will confirm the most deeply held religious beliefs of one religion. Which one? I guess they can fight a war over that to decide who wins.

Dembski’s methods of design detection can discriminate between informational patterns that are produced by chance/law, or alternatively were produced by intelligence. When there is real design to be detected, Dembski’s methods of design detection can work regardless of whether the designer was human or non-human. But Dr. Musgrave’s challenge seems to only allow intelligent design when it is human design. If we assume that Dr. Musgrave’s challenge was issued in good faith, then it seems the challenge has an inappropriate assumption: namely, that naturally occuring gene sequences were not designed. Dr. Musgrave may think that the correct “answer” is that only certain sequences were designed, because he knows they were designed by humans. But someone applying rigorous methods of design detection might find that other sequences were designed as well. Dr. Musgrave might then proclaim that the ID proponent is wrong, when in fact he is the one who is wrong because he assumed from the beginning that no naturally occurring gene sequence was designed. This is something to keep in mind if anyone submits analyses here.

I actually felt Ian's challenge was weaker than it could be, since ID doesn't claim to be able to tell two designers apart (that's what Multiple Designer Theory is for :) ), but once again... how totally unexpected that no ID person even attempted the challenge. It makes me laugh every time to watch them run and hide, and then it makes me sick that they get away with calling us the censors.

I call shenanigans. ID doesn't have any science and they have to stop claiming that it does.

By the way I have solutions to global pollution and world hunger. I can't show you them because environmentalists have expelled me, but they really exist, honest. Why is my name not in encylopaedias already? I should get the Nobel prize at least.

Mr. Christopher,

While I agree that taking down these IDiots would be akin to shooting fish in a barrel, I don’t think going on the offensive is the smart strategy. I think that with the advent of a strong internet, the easy ability to document these events wherein ID /Creationisnm just doesn’t have a clue, stands for itself. The fact that they have done nothing positive speaks volumes. Attacking a defenseless opponent has no point. Furthermore, the strength of science is that it is NOT decided by debate, but rather the outcome of facts and the explanatory and predictive power of the theories they support.

Yep, just as I expected. The ID position is that all life is designed, so naturally Luskin will bitch that Ian has assumed his conclusion - that SOME life was not designed. Luskin's conclusion that all life was designed is NOT an assumption, it's an axiom because it's stated in a source not identified.

After reading the Luskin quote on Mr_Christopher's post, I find it odd that the 'methods of design detection' cannot distinguish between human design and non human design... (I am assuming he means supernatural design, I can tell birds nest and bee hives are non human designs, or supernatural designs for that matter)

Does the fact that the explanatory filter cannot tell human design from supernatural design mean that (some) humans are equal or greater in intelligence than the (mythical) designer?

Flint: Yep, just as I expected. The ID position is that all life is designed, so naturally Luskin will bitch that Ian has assumed his conclusion - that SOME life was not designed. Luskin's conclusion that all life was designed is NOT an assumption, it's an axiom because it's stated in a source not identified.

hooligan wrote Mr. Christopher, While I agree that taking down these IDiots would be akin to shooting fish in a barrel, I don’t think going on the offensive is the smart strategy. I think that with the advent of a strong internet, the easy ability to document these events wherein ID /Creationisnm just doesn’t have a clue, stands for itself. The fact that they have done nothing positive speaks volumes. Attacking a defenseless opponent has no point. Furthermore, the strength of science is that it is NOT decided by debate, but rather the outcome of facts and the explanatory and predictive power of the theories they support.

Hooligan, not wanting to argue but I think this worth responding to. We agree that science is not decided by debate, but public opinion often time is and that is where science continues to lose. In fact if not for the Constitution our public science classes would be overwhelmed with "creationism science" and "intelligent design". The creationists publicly lie and distort the truth to win public support. John Q Public does not read scientific peer-reviewed literature. They harldy read Scientific American or National geographic for that matter. Everything they know tends to come from news agencies be in paper, radio/tv or the internet. The creationist POV tends to dominate all those venues. And John Q Public does not read PT. So, my take on it is we can continue to allow the creationists to lie and distort (and naively think people come to PT to sort the rubbish from the truth) or we can go on the offensive. This could also be a perfect venue to actually try and educate John Q Public. For instance John Q Public does not even know what "predictive power" means in the first place (and I wish PZ would have defined that in his "debate" with Simmons). Anyhow, not looking to argue, my children will never step foot in a public school room so I'm not personally losing much sleep over any of it. I just see a real opportunity for science to educate John Q Public while refuting the nonsense pushed by the creationists. Or science can sit by and continue to face these creationist frontal assaults (darwinism breeds nazi children, there are no transitional fossils, IDists are victims of a darwinian conspiracy, etc). I know that's a crock of shit but John Q Public does not and unless an effort is made to unmask the nonsense and educate John Q Public we cant continue to look forward to more nonsense like we're seeing in Florida. True, science is formulated in peer-review journals, predictive theories, and labs, etc, but public opinion is not. Just sayin'...

hooligan wrote Mr. Christopher, While I agree that taking down these IDiots would be akin to shooting fish in a barrel, I don’t think going on the offensive is the smart strategy. I think that with the advent of a strong internet, the easy ability to document these events wherein ID /Creationisnm just doesn’t have a clue, stands for itself. The fact that they have done nothing positive speaks volumes. Attacking a defenseless opponent has no point. Furthermore, the strength of science is that it is NOT decided by debate, but rather the outcome of facts and the explanatory and predictive power of the theories they support.

Casey wrote:

"Dembski’s methods of design detection can discriminate between informational patterns that are produced by chance/law, or alternatively were produced by intelligence. When there is real design to be detected, Dembski’s methods of design detection can work regardless of whether the designer was human or non-human."

Great. So here is the perfect opportunity for the ID crowd to demonstrate once and for all the importance of the explanatory filter and complex specified complexity. All they have to do is perform the calculation on the sequence they say that God designed and the one that Craig Venter designed and subtract to find the difference. That will show us exactly how much smarter Craig is than God. What proportion of the complexity did God design and what proportion did Craig design? Inquiring minds want to know. Of course, they will also have to explain why they have never done any calculations like this before.

No, seriously. If the filter cannot discriminate between different sources of complexity then what good is it? Even if it reliably detects design, that design could come from aliens or humans or time travelers or time traveling aliens or whatever. None of which does any good at all unless the source can be identified. None of which does anythiing at all to determine whether God did it or not. It's the same old problem that has always plagued ID - we know somebody did somethin somehow, but we don't know who, why, what or where.

we know somebody did somethin somehow, but we don’t know who, why, what or where.

Could the "but all life is designed" excuse be finessed by throwing a few randomly generated sequences into the mix? There's nothing in the wording of the challenge that requires all the sequences to have come from actual organisms.

I wonder if the possibilities (natural sequence / human-designed sequence) weren't too limited.

Maybe next time you could also include:

- Completely random string of nucleotides, generated by computer
- A stretch of junk DNA from existing organism

Would the ID proponents be able, say, to find a single real, natural sequence among nine random ones?

Post Hoc Super Bowl comfy chair analysis

Ignoring the confusion associated with the change introduced by Musgrave, the attendant cries of foul by Luskin and the underlying question they raise of whether sequences that are not human designed are in fact themselves designed (1). The question evolves into one which asks; is Dembski’s explanatory filter (EF) capable of distinguishing human designed sequences from other unknown designed sequences (as viewed from ID) or human designed sequences from sequences that are the result of natural processes (as viewed from an evolutionary biologists position)? At least 1 ID supporter claims this is not required of ID (2). Musgrave observes that no one attempted to apply the EF to the problem citing Wilkins and Elsberry critique of the EF as a possible reason for the lack of entries based on the EF (3). The most common reason given for not applying the EF to the problem is that not enough additional information is provided with the sequences or that the sequences were too short to be of any utility (4). Only 1 ID advocate attempted any analysis, solving the problem using standard techniques, but in addition quickly produces a number of alternative scenarios with no methodology to distinguish between his competing hypotheses (5). Curiously most of the suggested ID approaches rely on comparative techniques which have their underpinnings in evolutionary theory and this is the precise approach taken by the winners (6).

Since the usual comments appear of “let them show us how to do it”, it might be fruitful to try to apply the EF to the problem (7, 8). But the EF relies on a purely mathematical approach to design detection, it is an a priori probabilistic calculation and I succumb to a fatal flaw, I’m not a mathematician. But to be a useful tool for biologists the EF must be understood and applicable by non-mathematicians. In other words those slogging around in the mud playing with slimy squiggly things and who may be spending too much time in the noon day sun should be able to come back to the lab and plug in data and get reliable answers. So I’m left with hand waving like those at UD. I do observe that the sequences provided by Musgrave are of 2 classes those from Mycoplasma genetalium and those modified and watermarked by the Venter group.
From the Luskin ID perspective these represent: Designed versus double designed.
From and evolutionary biologist perspective: Naturally occurring versus modified. By including the Venter watermark Musgrave increased the probability that the EF would catch the sequence, contrary to any assertion that the designers’ watermark might be unrecognizable to humanity, since from the ID perspective 2 layers of design are included in the sequence and one of the agents of design is known. It is unclear why the EF has not been adapted to differentiate between sequences of at least these 2 types. This would seem one of the tests any developer of a new genomics tool would want to include in their validation stage before any claims of utility would be made.

Delta Pi Gamma (Scientia et Fermentum)

URL Tiny Bubbles

(1) www.uncommondescent.com/intelligent-design/ian-musgraves-intelligent-design-challenge/#comment-167730

(2) www.uncommondescent.com/intelligent-design/musgrave-addendum-to-intelligent-design-challenge/#comment-168574

(3) www.talkdesign.org/faqs/theftovertoil/theftovertoil.html

(4) www.uncommondescent.com/intelligent-design/intelligent-design-challenge-challenge/

(5) www.uncommondescent.com/intelligent-design/ian-musgraves-intelligent-design-challenge/#comment-167650, www.uncommondescent.com/intelligent-design/musgrave-addendum-to-intelligent-design-challenge/#comment-168681

(6) pandasthumb.org/archives/2008/01/take-the-intell.html#comment-141757

(7) www.uncommondescent.com/intelligent-design/intelligent-design-challenge-challenge/#comment-167779

(8) www.uncommondescent.com/intelligent-design/intelligent-design-challenge-challenge/#comment-167797

Completely random string of nucleotides, generated by computer

You don’t want completely random nucleotides. 3 out of 64 codons would then be stop codons. You would want to build up frequency tables from the nucleotide database and generate random sequences with the same frequency rates.

Isn't watermarking form of steganography which is included in list of ID's empirical research: http://www.researchintelligentdesign.org/wiki/Biosteganography, isn't this something that they are set out to do?

Irony?

You don’t want completely random nucleotides. 3 out of 64 codons would then be stop codons. You would want to build up frequency tables from the nucleotide database and generate random sequences with the same frequency rates.

string of random proteins

The absence of the explanatory filter from responses to this problem is part of a larger picture. DI associates wax over its virtues. On the other hand searches of scientific literature suggest that in practice the filter is useless:

http://www.geocities.com/lclane2/filter.html

How about reverse engineering a string of random proteins?

The absence of the explanatory filter from responses to this problem is part of a larger picture. DI associates wax over its virtues. On the other hand searches of scientific literature suggest that in practice the filter is useless:

The irony regarding the "explanatory filter" is that it doesn't offer any explanation; it's merely an excuse for *rejecting* established explanations in order to replace them with that special brand of ignorance that goes under the name "goddidit".

But Dr. Musgrave’s challenge seems to only allow intelligent design when it is human design. If we assume that Dr. Musgrave’s challenge was issued in good faith, then it seems the challenge has an inappropriate assumption: namely, that naturally occuring gene sequences were not designed.

— Luskin

I think much of this discussion is missing the point. The Explanatory Filter (EF) doesn't claim to be a start-to-finish algorithm where you just put the sequence into one end, and a result comes out of the other end. The EF claims to be able to make a design inference on the basis of a pattern found in a sequence, but first you have to find the pattern, and there you're on your own.

For example, Dembski's analysis of his Contact/SETI example starts with the observation that the signal contains a sequence of prime numbers. In the same way, anyone who wants to solve the challenge needs to start by finding the Venter watermark, whether they're then going to use the EF or not. And the methods (if any) used by ID advocates to find the watermark say nothing about the EF.

Once you've found the pattern, the EF just tells you to (1) calculate the probablility of getting that pattern (or anything like it) under each non-design hypothesis you can think of; (2) reject any hypothesis if the resulting probability is too small; and (3) if you've rejected all the available hypotheses, infer "design". I haven't attempted to read the relevant Uncommon Descent thread(s), but I doubt whether anyone has bothered to do this. In any case, as has been pointed out many times before, this is just a God-of-the-Gaps method. If you can't think of any hypothesis that confers a sufficiently large probability on the observed sequence, or if you can't calculate a probability (as in the case of the evolutionary explanation for the bacterial flagellum), Dembski tells you to infer design.

Ian's discussion of how the challenge winners identified the designed sequences stops at the point where they spotted the Venter watermark. It doesn't address the question of how they inferred design from the presence of the watermark. It may have been an obvious inference, but that begs the question of what the underlying logic of that obvious inference was. Despite Ian's implication that they didn't use "convoluted statistical measures", there may well have been some sort of probabilistic reasoning going on at a subconscious level. For example, they may have been intuitively reasoning on the basis that the probability of getting the watermarks given human design was far higher than the probability of getting those same sequences by any natural process they could imagine. Dembski claims that the logic of the EF underlies our ordinary intuitive inferences of design, so it doesn't necessarily require the conscious use of statistical measures. Perhaps the winners were using the EF without knowing it. I don't actually believe that--I'm just playing devil's advocate. My point is that Ian's post doesn't address the question.

Incidentally, it hardly needs saying but Casey Luskin's objection is absurd. If the objection were valid, then the EF wouldn't be able to distinguish between a blank page and one printed with a Shakespeare sonnet, since the paper was made by a human designer in both cases. Moreover, in his Caputo example Dembski tells us the EF can distinguish between a fair draw and Caputo cheating, despite the fact that the ballot devices were designed by humans regardless of whether there was cheating involved.

Incidentally, it hardly needs saying but Casey Luskin’s objection is absurd. If the objection were valid, then the EF wouldn’t be able to distinguish between a blank page and one printed with a Shakespeare sonnet, since the paper was made by a human designer in both cases.

No his objection isn't absurd at all. I think he's got it exactly right. His objection is indeed valid. :P

By the way, proponentsistsers think everything in the whole darned universe is designed. So I don't know why they would even think of proposing a design filter unless maybe they wanted to get people to agree that, yes, some things are designed, and then after that they can drive that wedge all the way home baby.

Richard Wein: I think much of this discussion is missing the point. The Explanatory Filter (EF) doesn't claim to be a start-to-finish algorithm where you just put the sequence into one end, and a result comes out of the other end. The EF claims to be able to make a design inference on the basis of a pattern found in a sequence, but first you have to find the pattern, and there you're on your own.

"That is, people took what they knew of human design and applied it to the problem at hand."

Isn't that exactly what Dembski et al do - thereby proving that things which are designed must have been designed by humans? LoL!

Ah, but that would make it a tautology, and a tautology aren’t science!

Not so dubious is the paper From Information Geometry to Newtonian Dynamics, by Ariel Caticha & Carlo Cafaro, in 27th Intern. Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, AIP Conf. Proc. 954, 2007.

No 'physical' postulates are required; Newtonian mechanics and the conservation of energy follow from the Maximum Entropy Principle, a variant of Bayesian reasoning.

(However, it seems most unlikely that quantum mechanics can be similarly derived.)

Newtonian mechanics and the conservation of energy follow from the Maximum Entropy Principle, a variant of Bayesian reasoning.

Perhaps the laws of physics are deeply geometrical because they are practical rules to process information about the world and geometry is the uniquely natural tool to do just that. This notion, that the laws of physics are not laws of nature but rules of inference, seems outrageous but deserves serious attention.

the laws of physics are not laws of nature but rules of inference,

Btw, I can immediately see why information can't be responsible for energy conservation - entropy can increase in closed systems.

the conserved current