Dembski's argument in Chicago -- New? Persuasive?

Posted 2 October 2014 by Joe Felsenstein

On August 14, William Dembski spoke at the Computations in Science Seminar at the University of Chicago. Was this a sign that Dembski's arguments for intelligent design were being taken seriously by computational scientists? Did he present new evidence? There was no new evidence, and the invitation seems to have come from Dembski's Ph.D. advisor Leo Kadanoff. I wasn't present, and you probably weren't either, but fortunately we can all view the seminar, as a video of it has been posted here on Youtube. It turns out that Dembski's current argument is based on two of his previous papers with Robert Marks (available here and here) so the arguments are not new. They involve considering a simple model of evolution in which we have all possible genotypes, each of which has a fitness. It's a simple model of evolution moving uphill on a fitness surface. Dembski and Marks argue that substantial evolutionary progress can only be made if the fitness surface is smooth enough, and that setting up a smooth enough fitness surface requires a Designer. Briefly, here's why I find their argument unconvincing:

They conside all possible ways that the set of fitnesses can be assigned to the set of genotypes. Almost all of these look like random assigments of fitnesses to genotypes.
Given that there is a random association of genotypes and fitnesses, Dembski is right to assert that it is very hard to make much progress in evolution. The fitness surface is a "white noise" surface that has a vast number of very sharp peaks. Evolution will make progress only until it climbs the nearest peak, and then it will stall. But ...
That is a very bad model for real biology, because in that case one mutation is as bad for you as changing all sites in your genome at the same time!
Also, in such a model all parts of the genome interact extremely strongly, much more than they do in real organisms.
Dembski and Marks acknowledge that if the fitness surface is smoother than that, progress can be made.
They then argue that choosing a smooth enough fitness surface out of all possible ways of associating the fitnesses with the genotypes requires a Designer.
But I argue that the ordinary laws of physics actually imply a surface a lot smoother than a random map of sequences to fitnesses. In particular if gene expression is separated in time and space, the genes are much less likely to interact strongly, and the fitness surface will be much smoother than the "white noise" surface.
Dembski and Marks implicitly acknowledge, though perhaps just for the sake of argument, that natural selection can create adaptation. Their argument does not require design to occur once the fitness surface is chosen. It is thus a Theistic Evolution argument rather than one that argues for Design Intervention.

That's a lot of argument to bite off in one chew. Let's go into more detail below the fold ... Dembski and Marks's argument involves defining a new form of information, showing that it is conserved. Evolution can succeed only if this information is already present, so therefore evolution does not bring about new information. In Dembski's case he goes on from that to make a theological argument (in his recent book), which I gather is basically "In the Beginning is the Information". People like to argue about how one ought to define information, but I'm going to ignore most of those arguments, because I think that there is a simpler problem that undercuts the Dembski-Marks argument. My argument here is not new (it has been given before at Panda's Thumb (here and also here and here). But with a new wave of publicity for Dembski and Marks's argument, it's worth pointing out in more detail the flaw in their argument. A typical fitness surface? Dembski and Marks have a simple model with genotypes and fitnesses. Of course it is overly simple, but all models are. It is worth examining, because if evolution is in trouble in such a model, we need to know why. Their computation of information is a measure of how smoothly the fitnesses change as one moves from one genotype to another, where neighboring genotypes are those that can be reached from each other by evolutionary processes such as mutation. If the fitness is smooth enough, one can find neighboring genotypes that are better, and the natural selection will tend to move the population to those. To figure out how common smooth fitness surfaces are, Dembski and Marks invoke Bernoulli's "Principal of Insufficient Reason". That basically says that if we can't think of a reason to consider probabilities of different outcomes unequal, we should consider the probabilities all to be equal. The use of Bernoulli's Principle underlies all of Dembski and Marks's calculations. In the case of fitness surfaces, the outcomes are all the different ways that fitnesses can be assigned to genotypes. So if we have DNA sequences, and the genome is 1000 bases long, there are 4 x 4 x 4 x ... x 4 different genotypes, with 1000 factors of 4 in the product. That is about 10 raised to the 602nd power. If each of these possible genotypes has a different fitness, there are also that many fitness values. The Principle of Insufficient Reason says that, lacking any reason to think otherwise, we should give each of the possible ways that the 10-to-the-602 fitnesses could be assigned to the genotypes an equal chance of being true. A typical one of this vast number of possibilities has fitnesses randomly assigned to genotypes. If that is the case, then when we change a genotype by making a single mutation in it, we arrive at a new genotype that has a fitness that is, in effect, chosen from all the possible fitnesses, at random. What a mutation does What if, instead of changing one base, we took the drastic step of mutating all of the bases in the genotype at the same time? If the Bernoulli Principle applied, we would get to a genotype whose fitness was also chosen at random. So in that case, on average, that would be no better and no worse than changing just one base. In other words, when fitnesses are randomly assigned to genotypes making a single typographical error is exactly as bad as changing every letter in the text . Real biology doesn't work anything like that. Making one mutation in one of my genes will on average make it worse, though sometimes not. If it produces a protein, a single amino acid change often leaves the protein still functioning. But making changes in every site of its DNA is the same as replacing every protein by a random string of amino acids. Which will be a complete disaster. Similarly, in statements in English, one typographical error might change "to be or not to be that is the question" into "to be or not to de that is the question". Changing all letters would give something like "bdglvwujzib lxmoxg rjdg a ohlowugrbl owj". It should be obvious that the latter is far less functional. The comprehensibility of English sentences is more like the actual fitness of organisms, and not like the fitness of the organisms Dembski and Marks imagine. Unbelievably strong interactions In Dembski and Marks's "white noise" fitness surfaces, another bizarre property is that every part of the genome interacts incredibly strongly with every other part. If they did not interact strongly, we would get cases like this: we might find that changing position number 834 in the DNA from C to T would make the genome somewhat worse. And changing position 95161 from A to C might also make the genome worse. If those two positions in the DNA underwent both of these changes at the same time, we could reasonably expect that this accumulation of two bad changes would be worse yet. But if the fitnesses are assigned to genotypes at random, that prediction could not be made. The double mutant would have a randomly-chosen fitness and that would have only about a 1/4 chance of being worse than either of the single mutants. In fact, about 1/4 of the time it would actually be better than the original genotype! We can immediately see that this could only happen if the two parts of the genome were tightly interacting in some way. But in the Dembski-Marks white noise model all parts of the genome interact tightly with each other. No real organism works that way. And there is a simple reason why. What physics does The reason is "because physics". In the physical laws of our universe, interaction at a distance gets weaker and weaker as the distance increases. This is an everyday fact that we rely on all the time. As I type these words my fingers and the keys move. There is an (extremely) slight gravitational and electrostatic effect of those movements on (say) the food in your refrigerator. That effect declines with distance. As a result, you don't have to worry that my typing is busy rearranging the food in your refrigerator. The eggs will be right where you left them, and this will not depend on whether I type the letter A or the letter B. Similarly, in the genome, a gene that functions in the growth of your toenails typically shows no strong interaction with another gene that controls nerve connections in your ear. They are physically far from each other and probably function at different times as well. In Dembski and Marks's model universe things don't work that way. If one particular gene has a mutant, we can't know anything about what its effect is, until we check all other genes. A change in any one of those others will make a major difference in what the effect of the first mutant is. And this is not just something that happens occasionally. It is always true, for all parts of the genome. Every gene, and every base in every gene, interacts incredibly tightly with all other bases in all other genes. Why the white-noise model prevents evolution The fitness surfaces implicit in Dembski and Marks's argument are known as "white noise" fitness surfaces. White noise has a signal whose values are uncorrelated from one time to another. The white noise fitness function is the same -- fitnesses of closely similar genotypes are totally dissimilar. Knowing the fitness of your genotype simply provides no prediction as to what the fitness will be if the base at one point in your DNA is changed. Natural selection with mutation and recombination can work its way uphill on the fitness surface by putting together individually favorable changes. If the fitness surface does not allow any prediction that such combinations will often be better than either change alone, then this is a big problem for evolution. The evolutionary process will frequently get stuck. Fortunately, "because physics" white noise fitness functions basically don't exist. What Dembski and Marks's argument doesn't do It is notable that Dembski and Marks's argument is not actually an Intelligent Design argument. It argues that a Designer is needed to explain the shape of the fitness surface, but once that surface is smooth enough, natural selection and other evolutionary forces do the rest. So there is no Design Intervention needed. Is evolution a search? Is it important whether it is? The audience at Dembski's talk in Chicago seemed to think that the crucial issue is whether evolution is or is not a "search". Strictly speaking, in a model of evolution like the one he is using, I think that the answer is no. But it actually is not important whether it is or isn't. Given the issue of whether a white noise fitness function is the default, Dembski's argument is invalid even if one allows him the point that evolution is a search. Has this criticism of Dembski's arguments been made before? Dembski also used a white noise fitness function in his No Free Lunch argument, and in the Search For a Search papers he and Marks acknowledge that connection. In the No Free Lunch argument the performance of the search that moves uphill on the fitness surface is extremely poor if averaged over all possible fitness functions. This is the same as its behavior on a typical randomly-chosen fitness function. At least seven major criticisms of Dembski's No Free Lunch argument have objected that white noise fitness functions are not realistic (links to their articles and posts are given in my 2007 article and in a summary I wrote here at Panda's Thumb). The criticism goes back to 2002 and has been voiced by all these authors. Dieb's argument with Dembski and Marks's theorem Mathematical blogger "Dieb" (Dietmar Eben) has raised the issue (here, here and here) of whether Marks and Dembski have actually proven their Horizontal No Free Lunch Theorem. His arguments are interesting and strike me as cogent. But whether or not that theorem is proven, the point remains that evolution will do badly almost all the time on a white noise fitness function. So a smoother fitness function is required. But, as we have seen, the laws of physics make a white noise fitness function unlikely. This is true whether or not the HNFL theorem can be proven rigorously. New types of information? Important to arguing for Design? The point about physics and the unlikelihood of white noise fitness functions is also true however we define information, and it is true whether natural selection "creates" information or whether it takes existing information that is implicit in the smoothness of the fitness surface and repackages it in the genome. I suspect that Dembski and Marks's "active information" will end up not being a helpful concept, but for the purposes of my present critique that issue is not central. What Richard Dawkins's "Weasel" model was not intended to do One should note in passing Dembski's use of Richard Dawkins's "Methinks It Is a Weasel" model. In his Chicago talk, Dembski portrays Dawkins as arguing that the Weasel model shows that natural selection can originate information, and portrays Dawkins as claiming that it is a realistic model of evolution. Dawkins was not arguing that it was a realistic model of evolution, or that this evolution originated new information. Dawkins's model was a teaching example to show why creationist debaters who argue that natural selection is doing a "random" search are disingenuous. The Weasel search succeeds in about 1000 steps, while a truly random search would take astronomical numbers of steps. Dawkins's model is an effective teaching device. It is routinely misrepresented in the creationist and ID literature as intended to be a realistic model of evolution, and intended to prove assertions about where the information in life originates. Unfortunately Dembski has followed this sad tradition. Is Dembski's theology of information central to his argument about evolution? No, because he's got to end up arguing that, for the laws of phyics to be the way they are, requires some active Design. But once the laws of physics are admitted, how they got that way is just not part of any argument about evolution. Biologists will certainly decide not to waste time on the issue and to leave it to cosmologists.

168 Comments

callahanpb · 2 October 2014

Dembski and Marks argue that substantial evolutionary progress can only be made if the fitness surface is smooth enough, and that setting up a smooth enough fitness surface requires a Designer.

"Come back here and take what's coming to you! I'll bite your legs off!" So it sounds like Dembski finally figured out that evolutionary programs can optimize things in a realistic-looking search space. Which in some sense is progress. It should be noted that their latest, last gasp is very far from Paley's original argument. If you accept that a sufficiently smooth search space can result in self-organized complexity, then the existence of self-organized complexity does not require a designer, contrary to Paley. The idea that smoothness requires a designer is just silly, but I guess you have to fight with the reality you have, not the reality you want.

eric · 2 October 2014

Given that there is a random association of genotypes and fitnesses, Dembski is right to assert that it is very hard to make much progress in evolution

In addition to the problems Joe mentions, there's another "unrealism" in this model. In real life, genotypes produce phenotypes, and a change in genotype will not change phenotype at all a large portion of the time. Any mutation to a different 3-letter group that codes for the same amino acid will have no effect (example: CCC to CCA). Any change to a non-coding region will likewise not have no effect. So the "fitness surface" for genotypes should consist of large flat regions separated by slopes (which may or may not be steep), not sharp peaks. Moreover, this has nothing to do with Dembski's contention that smooth surfaces must come from God. We simply observe that many changes in genotype produce no change in phenotype in fact.

eric · 2 October 2014

Now that I think about it, this random assignment of fitnesses means that Dembski is making one of the oldest and most naive mistakes in the creationist book, a mistake we commonly hear more from less sophisticated/educated creationists. He is thinking of evolution as a saltational process, where a relatively small change in code has a monkey produce a man or chicken come from a dinosaur. That is, in effect, what a random assignment of fitnesses to codes would model.

TomS · 2 October 2014

I wonder whether they tell what an intelligent designer does to shape a fitness landscape.

What do we know about Intelligent Designers that leads to there being one shape rather than another?

Might there be something, who knows what, that differs somehow from Intelligent Designers, that also leads to there being one shape rather than another?

How are designs - those concepts that Intelligent Designers design - implemented in the material of the natural world?

Is there anything else that the Intelligent Designers do, other than start off the process?

For those with theological interests, compare and contrast Intelligent Designers with the demiurge of Gnosticism, the god of pantheism or of deism, etc.

It may seem too much to ask for unconditional surrender, once there has been retreat on the issue of common descent with modification by natural means of much of the world of life (notably including humans) over billions of years. But it may be needed to advert any possibility of recrudescence.

Henry J · 2 October 2014

Joe,

How do you wade through that stuff in detail without flipping out?

It sounds like he's set up some sort of abstract mathematics that has no grounding in reality. (Kind of analogous to that guy does in one thread over on AtBC.)

On that business of computing "fitness" from the genotype, over all possible genotypes, the next question is - in what environment?. Fitness is after all relative to environment, and that includes what else is living in the same ecosystem.

Henry

lantog · 2 October 2014

Joe,

Thanks for the post. In places you and others use rather mild language to describe Dembski's work but overall I get the impression that he and Marks have failed spectacularly to produce anything of worth. I'm actually shocked. I would have thought that someone with a phd in math from the U of Chicago wouldn't waste almost a decade of effort on something so glaringly wrong, no matter how strong his religious bias. It seems like they made no effort at all to realistically consider fitness landscapes. So my question is; am I wrong in this impression? Is there something of worth in his efforts?

You mention you don't think of evolution as travel on a fitness landscape. Reading about this many years ago (S. Kaufmann?) I got the same impression. I don't think genotypes move on a landscape- they deform the landscape under them and fitness peaks are rare and are the low hanging fruit of evolutionary studies. Mostly living things evolve not to be dead. So if anything the landscape is flat with lots of pot-holes

DS · 2 October 2014

So have they ever tried to, you know, actually measure the fitness of any particular genotype? Do they always just make up the fitnesses and expect everyone to play along? Do they have any justification whatsoever for any of their assumptions about fitness, or is it all just a bunch of made up nonsense? Is the whole point that if the world were completely whacky in a special way that only they can imagine, that then evolution might not work, or do they have some other more rational point to make? Do they actually get paid for making up crap that sounds good but actually is a gross distortion of reality, or do they have to do some real science at at some point in order to get paid?

fnxtr · 2 October 2014

Adding to what Henry said, the environment isn't white noise, either.
Dembski is edging toward "Universe, therefore God". Which is just silly.

https://www.google.com/accounts/o8/id?id=AItOawnKupVGX70N9ZsvLu8iScIzWpyVj8bds_Q · 2 October 2014

Adding to what Henry said, the environment isnât white noise, either. Dembski is ~~edging toward âUniverse, therefore Godâ. Which is~~ just silly.

Better.

Rolf · 3 October 2014

Henry J said: Joe, How do you wade through that stuff in detail without flipping out? It sounds like he's set up some sort of abstract mathematics that has no grounding in reality. (Kind of analogous to that guy does in one thread over on AtBC.) On that business of computing "fitness" from the genotype, over all possible genotypes, the next question is - in what environment?. Fitness is after all relative to environment, and that includes what else is living in the same ecosystem. Henry

Thank you. I am very interested in this stuff but have no qualfifications. But while reading the OP I kept thinking just the same thing, a population is not reqired to stay in the same environment, not to mention that the environments where real life finds itself are not static. Plus, we already know that evolution is a fact.

https://www.google.com/accounts/o8/id?id=AItOawlcBX2t2NtC8gl0s6fe9Uvay2AD7GHQIMo · 3 October 2014

I think it is worth noting that this white noise model doesn't actually display any of the properties of genetics that we actually observe. Heredity does not exist. Children do not look like their parents, and siblings look no more similar than two randomly selected humans. Genetic diseases do not segregate within families, and cannot be tested for genetically. People from the same ethnic group do not look any more similar than those from different ethnic groups. Breeds do not breed true, and (taken to its conclusion) species do not exist. In general, individuals with more similar genomes do not have more similar phenotypes. A world where the map between genotype and phenotype is random is emphatically not the world we live in.

So basically this guy's argument is: "If you imagine a theoretical world, where genetics as we know it does not exist, then evolution cannot occur. The fact that we are in a world where genetics does exist must be because an unnamed Designer made the world like this to allow evolution to happen." Well ok then guy, good work.

https://www.google.com/accounts/o8/id?id=AItOawnwsK3wlP8NWGKEz-zi8I95cJLbCdN-pWQ · 3 October 2014

OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses?

What do you have besides bashing ID with misrepresentations?

harold · 3 October 2014

There are two simple ways to deal with this kind of crap.

Neither includes semantic wrangling over whether or not something is "random". A random variable exists when we can know the frequency at which different states can be expected to occur, but can't predict which will occur next. E.g. rolling a die. E.g. which mutations will occur when a genome replicates; mutation occurences are extremely well modeled as random variables; in fact if they weren't there would be no field of population genetics. You can make a good argument that "the environment" (local, global, universe, whatever) is well modeled as a random variable as well. Within bounds, it changes. Various states are more likely to occur than various other states but we can't predict perfectly which will occur next.

When creationists assign wrong probabilities to a random variable, that doesn't mean that the underlying concept of modelling something as a random variable is wrong.

However, there are some rather blatant problems with Dembski's nonsense.

1) Okay, so under Dembski's model, what happened, where, when? Who is the designer? What did the designer do? How did the designer do it? When did the designer do it? How can we test these answers? Etc.

2) Also, where is Dembski's fair discussion of the theory of evolution, why it seems to work, and precisely which problem his model solves better? That's how you advance science. Einstein didn't absurdly create a model that denies known physical observations and claims that Newtonian approximations don't work where they obviously do work. He expanded a working model to cover the extreme instances where it was not making correct predictions. Dembski is trying to "prove that evolution can't be possible", but we already know that the theory of evolution is supported by multiple converging lines of evidence and makes good predictions. He needs to deal with that before claiming that it is "impossible".

eric · 3 October 2014

https://www.google.com/accounts/o8/id?id=AItOawlcBX2t2NtC8gl0s6fe9Uvay2AD7GHQIMo said: So basically this guy's argument is: "If you imagine a theoretical world, where genetics as we know it does not exist, then evolution cannot occur. The fact that we are in a world where genetics does exist must be because an unnamed Designer made the world like this to allow evolution to happen." Well ok then guy, good work.

Yes, that's it in a nutshell. Behe does something similar but not as extreme; if you recall the published work he defended at Dover, that could be summarized as "imagine a world in which there is no exaptation. Now we will calculate probabilities of a multi-mutational benefit occurring in that world..." To be as charitable as possible and to answer lantog's implied question about worth, well...their math seems to be internally consistent. IOW they seem to be correctly drawing conclusions about the models they are testing. And many theoreticians in the past have contributed useful information to science by testing out simplified models of things. So their work could be useful in the future - if, for example, we discover some local phenomenon or ecology that has the properties of their model. But these models are not very useful for giving accurate predictions of the speed, likelihood, or capabilities of evolutionary processes in the standard biologial systems and ecologies we see around us.

https://www.google.com/accounts/o8/id?id=AItOawnwsK3wlP8NWGKEz-zi8I95cJLbCdN-pWQ said: OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses? What do you have besides bashing ID with misrepresentations?

Fossil evidence of descent with modification. A common genetic code for all organisms. Multiple, independent cladistics methods that give basically the same results, and which match what we see in the genetic code. Observed speciation. Observation of mutations across generations, yielding novel phenotypic traits. Models of genetic drift. And that was what I could find in about two minutes. It literally took me longer to compose this message than it took me to find the information you think science doesn't have. It has that info in volumes, running to thousands or tens of thousands of pages. All you have to do is look.

DS · 3 October 2014

https://www.google.com/accounts/o8/id?id=AItOawnwsK3wlP8NWGKEz-zi8I95cJLbCdN-pWQ said: OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses? What do you have besides bashing ID with misrepresentations?

I think you miss the point. Dembski is making up crap in order to misrepresent evolution. The point is to show him up for the disingenuous an fundamentally dishonest charlatan that he is. There are many models that show that evolution is possible of producing the diversity of life that we observe, but that is not the point. Dembski ignores all of these without even trying to address them. That is why everyone can easily see through his transparent charade. And of course, even if no such models existed, as eric points out, we would still know that evolution was capable of producing the diversity of life because of the evidence that it actually has done so.

DiEb · 3 October 2014

For the moment, I just want to plug shamelessly my blog, where I tried to make a transcript of the video: William Dembski's talk at the University of Chicago, garnished with some annotations.

Joe Felsenstein · 3 October 2014

callahanpb said: ... So it sounds like Dembski finally figured out that evolutionary programs can optimize things in a realistic-looking search space. Which in some sense is progress. It should be noted that their latest, last gasp is very far from Paley's original argument. If you accept that a sufficiently smooth search space can result in self-organized complexity, then the existence of self-organized complexity does not require a designer, contrary to Paley. The idea that smoothness requires a designer is just silly, but I guess you have to fight with the reality you have, not the reality you want.

I just wish I knew what "self-organized" meant. However if you replace "self-organized complexity" with "complex adaptations produced by natural selection" there is no disagreement between us.

Joe Felsenstein · 3 October 2014

eric said: Now that I think about it, this random assignment of fitnesses means that Dembski is making one of the oldest and most naive mistakes in the creationist book, a mistake we commonly hear more from less sophisticated/educated creationists. He is thinking of evolution as a saltational process, where a relatively small change in code has a monkey produce a man or chicken come from a dinosaur. That is, in effect, what a random assignment of fitnesses to codes would model.

I'm not so sure that this is the source of his model. In the white noise fitness surfaces, almost any change will produce, not a dramatically different well-adapted organism, but a dead one. You could climb very local peaks, but just for a modest number of steps, and then you would stall out. The issue of how phenotype would change at each adaptive step is interesting -- yes, probably big changes at each step. But I think that this was not why the surface is used. It is just to show that realistically smooth ones, that do not involve many saltations, require Design Intervention (at least at the beginning of the process).

Scott F · 3 October 2014

eric said: Now that I think about it, this random assignment of fitnesses means that Dembski is making one of the oldest and most naive mistakes in the creationist book, a mistake we commonly hear more from less sophisticated/educated creationists. He is thinking of evolution as a saltational process, where a relatively small change in code has a monkey produce a man or chicken come from a dinosaur. That is, in effect, what a random assignment of fitnesses to codes would model.

Don't we see both kinds of mutation, with both minimal and large effects on phenotype? That is, small point changes typically don't have much effect, but occasionally relatively small changes in certain regulatory regions can have dramatic effects, such as adding a whole new body segment in insects, or a sixth finger or extra (or fewer) wisdom teeth in humans.

Joe Felsenstein · 3 October 2014

lantog said: ... You mention you don't think of evolution as travel on a fitness landscape. Reading about this many years ago (S. Kaufmann?) I got the same impression. I don't think genotypes move on a landscape- they deform the landscape under them and fitness peaks are rare and are the low hanging fruit of evolutionary studies. Mostly living things evolve not to be dead. So if anything the landscape is flat with lots of pot-holes

One can make more sophisticated and more realistic models of evolution, as you have done. But one can't do much math with them. Fitness surfaces (adaptive topographies, adaptive landscapes, etc.) are useful as models, and no, I have not stopped using them. The Evolutionary Quantitative Genetics tutorial that Steve Arnold and I teach each summer is full of them. The use of simple models in this discussion is to see whether, even in those simple cases, there is a problem with the ability of natural selection to increase adaptation. At first Dembski's Design Inference argument seemed to be saying that. Later he clarified that he was not claiming that. A claim that simple models won't produce adaptation seemed to be a bad bet, because if these models did not produce adaptation, you'd think that RA Fisher, Sewall Wright, and JBS Haldane would have noticed. You had to get up very very early in the morning to outthink them.

TomS · 3 October 2014

harold said:T 1) Okay, so under Dembski's model, what happened, where, when? Who is the designer? What did the designer do? How did the designer do it? When did the designer do it? How can we test these answers? Etc. 2) Also, where is Dembski's fair discussion of the theory of evolution, why it seems to work, and precisely which problem his model solves better? That's how you advance science. Einstein didn't absurdly create a model that denies known physical observations and claims that Newtonian approximations don't work where they obviously do work. He expanded a working model to cover the extreme instances where it was not making correct predictions. Dembski is trying to "prove that evolution can't be possible", but we already know that the theory of evolution is supported by multiple converging lines of evidence and makes good predictions. He needs to deal with that before claiming that it is "impossible".

That bears keeping in mind whatever "new" comes from those who voice objections to standard biological science. It isn't even a matter of science that we're talking about. Let's say we want to know why Hamlet says "To be or not to be." First of all, it isn't very helpful to say that Shakespeare wrote it. Even though it's true, it doesn't answer the question. But it might be part of an explanation involving when Shakespeare lived, where he lived, what other things he wrote. So it does do a little bit. I'm not so informed about literature, but probably some Shakespeare scholar could tell me why. Or maybe Freud would have his way of explaining those words. But if someone says, "Shakespeare couldn't have written it, so there must be an intelligent playwright who did write it" - what is the first thing that you think of? "Who did write it? Tell me something about this playwright. Or playwrights. Was it Francis Bacon, or the Earl of Oxford, or even some Klingon? And, by the way, how do you account for this phrase?" I will not be satisfied to know that intelligent playwrights are up to the task of writing just about anything, so an IP could have written "To be or not to be". No matter how convincing a case that you have for the impossibility of Shakespeare having written it, you haven't answered the question about that particular phrase.

harold · 3 October 2014

OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses?

The evidence is basically all of mainstream biomedical science. Molecular genetics, biochemistry, cell biology, anatomy, physiology, and paleontology all provide literally thousands of strong independent lines of evidence that support the theory of evolution. They all do so in ways that are compatible with each other and with physics and chemistry. Each time a new biomedical advance has been made since at least the nineteenth century, it has further supported and clarified the role of evolution in biology. I assure you that this has nothing to do with my personal wishes. It's simply the way the evidence shows that the world works, to any reasonable observer. But let me ask you a related question - Is there any evidence that would convince you, and if so, what evidence for evolution, now lacking, would do so?

What do you have besides bashing ID with misrepresentations?

You claim there are misrepresentations but don't point out even a single example. This makes it look as if you saw a fair critique, didn't like it, and childishly declared it to be a "misrepresentation". You can, however, improve this first impression by giving an example, from either the article or the comments section, of an actual misrepresentation, along with a cogent explanation of why it is so. Also, could you please answer the following questions - 1) Who is the designer? 2) What did the designer do? 3) When did the designer do it? 4) How did the designer do it? 5) Why did the designer make it "look like" evolution? 6) How can we test your answers? Also, one more question - Imagine a hypothetical state without First Amendment restrictions. Public school curriculum can be decided entirely by popular vote, without limitations based on either expert opinion or the constitution. In this state, there is a controversy. Some people want to teach Dembski-style ID as science, but others wish to teach outright YEC creation science, with specific reference to a young earth, a Garden of Eden, Adam and Eve, the Noahic flood, Jonah in the whale, etc, as literally true scientific fact. Which of these do you favor, and why?

Joe Felsenstein · 3 October 2014

DS said: So have they ever tried to, you know, actually measure the fitness of any particular genotype? Do they always just make up the fitnesses and expect everyone to play along? Do they have any justification whatsoever for any of their assumptions about fitness, or is it all just a bunch of made up nonsense? ...

Their point is that among all possible fitness functions, the ones where evolution succeeds are rare and need to be found by a Designer. It's just that the distribution that they are using ignores physics. I wince a little when you accuse them of not ever having measured fitnesses of real organisms. After all, I haven't either. Their argument tries to make a general statement that holds across all possibilities so that we can say something without having available any large generalities about fitnesses. That mode of operation is what theoretical population genetics used to use when I trained in the field 50 years ago. Now, since the availability of molecular data and since the development of more intensive efforts to measure fitnesses in various interesting ways, we are moving away from that, But it is slow going, since we can only measure largish differences in fitness, and the population can respond to much smaller ones.

Joe Felsenstein · 3 October 2014

https://www.google.com/accounts/o8/id?id=AItOawnwsK3wlP8NWGKEz-zi8I95cJLbCdN-pWQ said: OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses? What do you have besides bashing ID with misrepresentations?

If this particular troll has any evidence that I have misrepresented Dembski and Marks's argument, they should present it here. That would be on-topic. Otherwise the question is intelligently designed to derail the discussion into an endless roiling discussion in which the troll demands a complete model of everything, is never satisfied with the answers, and (of course) declares victory. I am going to patro[l]l this thread as aggressively as I can and send off-topic stuff to the Bathroom Wall. That includes all troll-chasing. You have all been warned.

Joe Felsenstein · 3 October 2014

DiEb said: For the moment, I just want to plug shamelessly my blog, where I tried to make a transcript of the video: William Dembski's talk at the University of Chicago, garnished with some annotations.

I gave some links to other stuff on your blog, namely the technical arguments that Dembski and Marks's Horizontal No Free Lunch Theorem has not been proven. I found the video fairly clear, though not professionally recorded. But your blog transcript will be a useful place to pick up quotes from Dembski's talk, plus your take on what the problems in his argument are.

Joe Felsenstein · 3 October 2014

harold said:
[-pwQ:] OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses?
The evidence is basically all of mainstream biomedical science. ... [snipped]

Just to repeat -- chasing this argument of this troll will henceforth occur on the Bathroom Wall. ...

[harold:]
[-pwQ:] What do you have besides bashing ID with misrepresentations?
You claim there are misrepresentations but don't point out even a single example. This makes it look as if you saw a fair critique, didn't like it, and childishly declared it to be a "misrepresentation". You can, however, improve this first impression by giving an example, from either the article or the comments section, of an actual misrepresentation, along with a cogent explanation of why it is so.

I agree with this (on-topic) comment by harold. The rest of harold's comment is chasing off-topic trollery, and all that will in future go to the BW.

Joe Felsenstein · 3 October 2014

eric said: [Interesting and relevant stuff snipped -- JF]
https://www.google.com/accounts/o8/id?id=AItOawnwsK3wlP8NWGKEz-zi8I95cJLbCdN-pWQ said: OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses? What do you have besides bashing ID with misrepresentations?
Fossil evidence of descent with modification. A common genetic code for all organisms. Multiple, independent cladistics methods ... [more snipping -- JF]

Again a warning -- chasing the troll -pwQ's "what is the evidence ... Where are the models?" argument will in the future go to the Bathroom Wall. The accusation by -pwQ of misrepresentation is on-topic and discussion of it will be allowed.

ceplaw · 3 October 2014

There are two other huge, unstated assumptions in the Dembski/Marks model that are both pretty easily refuted:

(1) The purported "fitness surface" is a uniform-density, omnidirectional gradient that operates simultaneously in all directions. This is simpler to describe mathematically than a nonuniform or direction-biased gradient... but is inconsistent with the second and third laws of thermodynamics (not to mention the common sense observation that it's harder to fix a broken simple machine, such as a lever, than it is to break it).

We'll leave aside for the moment whether a "surface" is a valid mathematical representation of a system that includes both Hermetian and non-Hermetian elements.

(2) Only the net energy cost of a transformation of any kind matters in any sense. The chemical physics refutation is the concept of activation energy... which is confirmed rather emphatically through the action of enzymes, and points out that overall efficiency may not be the ~~Holy Grail~~ ~~objective~~ life-affirming pathway.

For example, the most-efficient way of obtaining energy from, say, simple sugars is simple combustion. That, however, requires a combustion source (usually not friendly to or perhaps even possible for microorganisms living in watery environments)... and produces certain waste products, such as a sudden heat spike, rather inimical to life as we know it. It's actually fairly easy to determine the input energy and time period necessary to burn glucose in a lab. All of those safety precautions one must take in the lab should be a hint that a necessary precondition for life is that there be an alternative to simple combustion to release stored energy... and that that alternative will not be thermodynamically efficient, but will instead require a number of kludges. (Really: ATP/ADP and the Krebs Cycle?)

Henry J · 3 October 2014

Weâll leave aside for the moment whether a âsurfaceâ is a valid mathematical representation of a system that includes both Hermetian and non-Hermetian elements.

Not to mention the huge number of dimensions of this so-called "surface".

harold · 3 October 2014

In case that particular Masked Panda wishes to reply to my general queries, I have copied and pasted that part of my comment to the BW. I probably won't have time to keep up with the BW until later, but I'm sure others may be interested in pursuing any line of discussion that ensues.

http://pandasthumb.org/bw/#comment-334106

DS · 3 October 2014

Joe Felsenstein said:
DS said: So have they ever tried to, you know, actually measure the fitness of any particular genotype? Do they always just make up the fitnesses and expect everyone to play along? Do they have any justification whatsoever for any of their assumptions about fitness, or is it all just a bunch of made up nonsense? ...
Their point is that among all possible fitness functions, the ones where evolution succeeds are rare and need to be found by a Designer. It's just that the distribution that they are using ignores physics. I wince a little when you accuse them of not ever having measured fitnesses of real organisms. After all, I haven't either. Their argument tries to make a general statement that holds across all possibilities so that we can say something without having available any large generalities about fitnesses. That mode of operation is what theoretical population genetics used to use when I trained in the field 50 years ago. Now, since the availability of molecular data and since the development of more intensive efforts to measure fitnesses in various interesting ways, we are moving away from that, But it is slow going, since we can only measure largish differences in fitness, and the population can respond to much smaller ones.

I understand that it is a theoretical argument. However, if it does not apply to the real world, it is worthless. It is not adequate to envision some circumstances under which evolution might not work. It is necessary to demonstrate that it cannot work under realistic conditions. If one never measures any fitness values or has any idea what he real fitness landscape looks like, then it seems counter productive to try to infer that evolution cannot happen. And it doesn't matter if the conditions required for evolution to work are rare if they actually exist. That's just another bad application of the anthropic principle.

Joe Felsenstein · 3 October 2014

Henry J said:
Weâll leave aside for the moment whether a âsurfaceâ is a valid mathematical representation of a system that includes both Hermetian and non-Hermetian elements.
Not to mention the huge number of dimensions of this so-called "surface".

The high numbrt of dimensions does, as you note, mean that thinking of it as a "surface" is not a very precise analogy. Dembski and Marks are not the only people that have that problem -- population genetics and quantitative genetics do too. But mathematical results can be obtained anyway. PS I think it must be "Hermitian" after the mathematician Hermite. But I still have not entirely figured out what commenter "ceplaw" is actually saying.

Joe Felsenstein · 3 October 2014

DS said: ... I understand that it is a theoretical argument. However, if it does not apply to the real world, it is worthless. It is not adequate to envision some circumstances under which evolution might not work. It is necessary to demonstrate that it cannot work under realistic conditions. If one never measures any fitness values or has any idea what he real fitness landscape looks like, then it seems counter productive to try to infer that evolution cannot happen. And it doesn't matter if the conditions required for evolution to work are rare if they actually exist. That's just another bad application of the anthropic principle.

Dembski and Marks are not saying evolution can't happen. They are saying that even if it can, a Designer had to carry out a Design Intervention to set up the fitness surface so that it is the right shape for evolution to succeed. They're wrong "because phyiscs", but that's what they're saying.

TomS · 3 October 2014

harold said:
OK but what is the evidence that natural selection, drift and/or neutral substitutions are up to the task? Where are the models? Where are the testable hypotheses?
The evidence is basically all of mainstream biomedical science. Molecular genetics, biochemistry, cell biology, anatomy, physiology, and paleontology all provide literally thousands of strong independent lines of evidence that support the theory of evolution. They all do so in ways that are compatible with each other and with physics and chemistry. Each time a new biomedical advance has been made since at least the nineteenth century, it has further supported and clarified the role of evolution in biology.

You forgot biogeography and embryology. And what makes it interesting is that the existence of creationism, and particularly ID gives testimony to the success of evolutionary biology. For ID, because it does not even attempt to address any of the features of life that evolutionary biology elucidates, calls attention to the fact that no one has offered an description of life on Earth which accounts for the variety of life without common descent and modification - except the idea that life as existed as it is from eternity (or somehow in endless cycles), or the Omphalos hypothesis (that things, when they first appeared, were as if they had a prior existence). (Yes, there are alternative mechanisms for the descent with modification. Not only "natural selection". That's a matter of details which scientists are continuing to work on.) If they had any hypothesis, we would expect to hear from the anti-evolutionists, given that all of those clever people have had well over a century to work on what that alternative might be like. As far as I can tell from the latest, Dembski, as well as others, is engaged in what may be called "unintentional apophasis" (I first came across the phrase in the Wikipedia article on "The lady doth protest too much, methinks").

https://me.yahoo.com/a/JxVN0eQFqtmgoY7wC1cZM44ET_iAanxHQmLgYgX_Zhn8#57cad · 3 October 2014

Evolution will make progress only until it climbs the nearest peak, and then it will stall.

They seem quite aware of how evolution can end up stranded on a peak. But they will never deal with the fact that life has endless numbers of features that have done just that. Not stalled, so much as not being able to go back down and climb a better peak, as design can clearly do. So feathers are almost impossible for bats to evolve, while birds will never evolve the three ossicle ears of mammals. Thus what we see is what evolution predicts, mammals climb mount fur/hair, with feathers never being a possibility, and likewise for the rest of the traits in various lineages (hence cladistics). But hey, why bother with the fact that evolution explains what we see in life, while design doesn't, when you can complain that evolution simply couldn't do it? Nothing but evolution produces what we see in life--definitely not intelligence--yet they can forever ignore that fact without coming up with anything like as good an explanation for the one-way paths of evolving life. Glen Davidson

ksplawn · 3 October 2014

So basically they're using a model of a system that turns out not to have any of the defining features of the real system they're modeling. Genetic changes have no variation in consequence and no relation to each other. If I have that right, it sounds an awful lot like the classic Creationist argument against abiogenesis, usually formulated: "If you have an assortment of atoms and let them interact by random chance, the probability of them producing a complex organic molecule by randomly coming together are beyond astronomical, therefore abiogenesis is virtually impossible!"* Which ignores the fact that in the real world, atoms do NOT interact in a random fashion. Each element's atom has characteristic behaviors that govern their interaction with other atoms, promoting some combinations and excluding others. Real chemistry doesn't behave anything like the Creationist's model of chemistry for abiogenesis.

callahanpb said: "Come back here and take what's coming to you! I'll bite your legs off!" So it sounds like Dembski finally figured out that evolutionary programs can optimize things in a realistic-looking search space. Which in some sense is progress.

Which is probably why they are unwilling to publish a study on evolutionary programs using a realistic-looking search space, and indeed must now insist that the realistic-looking search spaces are not realistic. Not unless, apparently, chosen by a Designer! Because only a Designer can pick out a realistic search-space from all possible search spaces. At first I thought this approach to arguing fitness surfaces was simply a version of the Anthropic Principle: the fitness surface was chosen for our benefit. But the better answer to the AP is that we exist because the fitness surface allows us to: in any world with a "white noise" fitness surface there couldn't be any people around to question their own existence. But then I started to think... how could any physical world be represented by a white noise fitness surface? Wouldn't that require the same mistake about random interactions with equal consequences among things described above? Even if you changed some fundamental constants of the Universe so that life as we know it was not possible, that Universe itself would still have distinct physical things with distinct physical properties that must interact with each other according to some non-random organizing system driven by their distinctiveness from each other. If there was any kind of chemistry in such a Universe, it must still follow some kind systematic governance of interactions. If there was any kind of life there, it would still be governed by the same overall principle of random information generation and selective filter as long as imperfect replication of its information was true, and we have no reason to think that it wouldn't be (even in a Universe with different physical constants). So I don't think even the Anthropic Principle helps here. A Universe where all interactions are equally probable and have equal consequences doesn't seem to describe any sort of physical world. I may be wrong about that. *It's been a long time since I've actually run up against these arguments in the real world, so I refreshed my memory on a few Creationist sites to make sure I didn't mischaracterize it. I didn't.

TomS · 3 October 2014

Joe Felsenstein said: Dembski and Marks are not saying evolution can't happen. They are saying that even if it can, a Designer had to carry out a Design Intervention to set up the fitness surface so that it is the right shape for evolution to succeed. They're wrong "because phyiscs", but that's what they're saying.

And we should graciously accept their surrender? Although the theologians may want to protest something which, in all of its vagueness, looks something like Gnosticism, Pantheism or Deism. Some kind of agency concerned with something or other. Whatever, not much like the God Who is in a personal relationship with me as Creator, Sustainer and Redeemer.

callahanpb · 3 October 2014

Joe Felsenstein said: I just wish I knew what "self-organized" meant. However if you replace "self-organized complexity" with "complex adaptations produced by natural selection" there is no disagreement between us.

Self-organization is not something I made up. See http://en.wikipedia.org/wiki/Self-organization I agree that the term "self-organized complexity" is probably ill-defined, mainly because "complexity" is ill-defined, so I concede that I wrote sloppily. Dembski's original claims (e.g. misuse of No Free Lunch) seem at first to deny the existence not only of complex adaptations produced by natural selection, but nearly any process that produces results that are not obviously encoded in the process. Since I'm not an evolutionary biologist, but am very familiar with self-organization in cellular automata, I tend to look at his claims in that context, where they are also obviously wrong, or at least misapplied. Clearly, you need a somewhat well-behaved search space to do any better than random probing, so throwing away that assumption or claiming it requires a designer is an act of desperation.

Carl Drews · 3 October 2014

Search algorithm:

I don't see what's wrong with comparing evolution to a search algorithm. I have used that analogy for people who think that a random process cannot get anywhere; natural selection allows a species to explore the environment for better fitness, and it's very non-random.

Designed fitness surface:

When the Chicxulub meteorite hit the Yucatan 65 million years ago, the natural environment suddenly changed. Mammals, who had found a local fitness peak in underground burrows (not very impressive), suddenly became the "top dogs" and filled the biological space formerly occupied by the dinosaurs. Standard evolutionary theory.

I don't see how the Intelligent Designer is supposed to have designed the mammalian (genetic?) fitness surface to anticipate that an impact event was coming. Did ID send them underground in anticipation of the blast?

TomS · 3 October 2014

Carl Drews said: I don't see how the Intelligent Designer is supposed to have designed the mammalian (genetic?) fitness surface to anticipate that an impact event was coming. Did ID send them underground in anticipation of the blast?

Did one of the IDs send the impactor? One presumably different from the ID which was responsible for the dinosaurs? Unless it was all part of plan for the diversification of the birds (and the mammals were unanticipated beneficiaries)?

Henry J · 3 October 2014

I donât see whatâs wrong with comparing evolution to a search algorithm.

Doesn't that phrase usually imply a search with one particular goal? I mean something more specific than simply finding a way of producing more offspring. Even if each particular species might be said to be searching for some way of improving its ways of finding food, avoiding becoming food, finding shelter, finding a mate, etc., that's still going to be different for different species.

Henry J · 3 October 2014

Did one of the IDs send the impactor? One presumably different from the ID which was responsible for the dinosaurs? Unless it was all part of plan for the diversification of the birds (and the mammals were unanticipated beneficiaries)?

Not to mention the one that let the atmosphere get cluttered up with oxygen, killing off (or driving underground) everybody who was adapted to air that didn't have that stuff in it.

Carl Drews · 3 October 2014

TomS said:
Carl Drews said: I don't see how the Intelligent Designer is supposed to have designed the mammalian (genetic?) fitness surface to anticipate that an impact event was coming. Did ID send them underground in anticipation of the blast?
Did one of the IDs send the impactor? One presumably different from the ID which was responsible for the dinosaurs? Unless it was all part of plan for the diversification of the birds (and the mammals were unanticipated beneficiaries)?

Yet again, the Intelligent Designer is trying to make it look like life evolved. ;-) Or maybe TomS' warring IDs are like the Olympian gods during the Trojan War, each rooting for their own favorite. Back to polytheism. Let's try a science question: Single biological lineages are known to have split and diversified. At what point does the Intelligent Designer make that happen? When does one genetic fitness surface get redesigned into two?

callahanpb · 3 October 2014

Henry J said:
I donât see what's wrong with comparing evolution to a search algorithm.
Doesn't that phrase usually imply a search with one particular goal? I mean something more specific than simply finding a way of producing more offspring. Even if each particular species might be said to be searching for some way of improving its ways of finding food, avoiding becoming food, finding shelter, finding a mate, etc., that's still going to be different for different species.

A search is generally defined in terms of satisfying constraints, rather than one particular goal. E.g., you can search for a way to pack a set of non-overlapping polygons into a particular area. There may be multiple goals but any one of them will suffice, and the algorithm used to find one is conventionally called a search. However, I agree that evolution also lacks an a priori set of constraints. I am not sure this is very important, because Dembski's arguments are equally inapplicable to evolution and to many kinds of search, often for similar reasons. Biologists are understandably uncomfortable to have people study, e.g. genetic algorithms, and mistakenly conclude that they know something about evolution. But there is still a useful overlap (in my opinion) provided you are aware of the limitations.

eric · 3 October 2014

Joe Felsenstein said: Dembski and Marks are not saying evolution can't happen. They are saying that even if it can, a Designer had to carry out a Design Intervention to set up the fitness surface so that it is the right shape for evolution to succeed. They're wrong "because phyiscs", but that's what they're saying.

I think part of their strategy here is to make a silly argument opaque by using scientific terminology. Saying that smooth fitness surfaces are so rare among the types of surfaces that their presence must be explained by design is arcane enough that it sounds credible. After all, unless you work with this sort of mathematical calucation, how can you really refute that? Its hard to even picture what they're talking about - the set of all possible fitness landscapes, which are not two- or even three- dimensional, but many-dimensional, and the percent of them which are smooth by some criteria. But now let's translate that general, arcane claim into what it means for concrete genetics. There are 64 possible 3-letter combinations of our base units. These code for 21 amino acids and a "stop." That means that 42 of the combinations are redundant. In terms of landscapes, that is pretty darn flat; it means that most single point mutations will have no effect at all. And Dembski's claim amounts to: "redundancies in the number of unique codons per amino acid/stop are so unlikely to arise naturally that they can only be explained by God." Phrased in that more simple and concrete manner, it's patently ridiculous, isn't it? And that is just one aspect of the 'because physics' response. There are many many more ways in which having a backdrop of natural processes could be expected to lead to smooth fitness surfaces. Smooth is not unusual. Quite the opposite, in fact: smooth fitness landscapes are generally what you should expect to arise in/from nonchaotic systems.

callahanpb · 3 October 2014

eric said: Its hard to even picture what they're talking about - the set of all possible fitness landscapes, which are not two- or even three- dimensional, but many-dimensional, and the percent of them which are smooth by some criteria.

Maybe I'm being unfair, but it certainly does seem silly. E.g., the number 1 has a lot of useful properties, such as being the multiplicative identity. But the probability of choosing 1 randomly from the set of possible integers is 0. Therefore the number 1 must be designed. The unit circle, also very nice, has probability 0 of being chosen uniformly from the set of all closed curves in the plane (let alone the set of all point sets). Therefore circles must have a designer. Unless I am missing something subtle, it seems like Dembski is grasping at the very straws generally assumed as axioms to claim that there is a designer. Is there anything his argument could not apply to?

Joe Felsenstein · 3 October 2014

ksplawn said: So basically they're using a model of a system that turns out not to have any of the defining features of the real system they're modeling. Genetic changes have no variation in consequence and no relation to each other.

No, in their fitness surfaces genetic changes are overwhelmingly disastrous.

... But then I started to think... how could any physical world be represented by a white noise fitness surface? Wouldn't that require the same mistake about random interactions with equal consequences among things described above? Even if you changed some fundamental constants of the Universe so that life as we know it was not possible, that Universe itself would still have distinct physical things with distinct physical properties that must interact with each other according to some non-random organizing system driven by their distinctiveness from each other. If there was any kind of chemistry in such a Universe, it must still follow some kind systematic governance of interactions. If there was any kind of life there, it would still be governed by the same overall principle of random information generation and selective filter as long as imperfect replication of its information was true, and we have no reason to think that it wouldn't be (even in a Universe with different physical constants). So I don't think even the Anthropic Principle helps here. A Universe where all interactions are equally probable and have equal consequences doesn't seem to describe any sort of physical world. I may be wrong about that.

You're right -- it is very very hard to set up any physically realizable system that has that much interaction among its parts. The exception would be installing a cryptographic system between the genotype and the phenotypes. If the (haploid) genotypes were binary strings such as 100101110 ... 010111 and then one encrypted them, and got out a dramatically different string such as 0011101101010100 ... and took the individual digits as the phenotypes of different characters, and had the fitness be some reasonable function of the phenotype values, then the fitness surface as a function of the genotype string would be of the white-noise variety. By the way, I always thought that the Anthropic Principle was just that the properties of our part of the Universe may be different than average, simply because it has to be one in which we can arise and live. (I see from Wikipedia that this is actually the Weak Anthropic Principle).

TomS · 3 October 2014

I am not a scientist, but I can understand you smart, dedicated, informed and interested people love to tell us about science, and feel compelled to point out glaring errors in presentations of creationists. And we non-scientists want to hear more about real science. And it's a lot of fun for all of us.

But may I be permitted to note that ID and its companions don't have a chance of competing with science. Because they have decided that they don't stand a chance, and have given up, no more trying to present an account for "what happened and when". Their only hope is that nobody notices.

And, unfortunately, if they can engage in discussions about science which are difficult for us lay people to follow, that can, they hope, stave off the collapse of creationism. And I don't think that many of you experts realize that what you are saying is not obvious.

Anyway, I hope that you don't think that we lay people don't appreciate what you are doing.

It is fun.

Mike Elzinga · 3 October 2014

My general approach to any ID/creationist "paper" is not to try to follow or "understand" their "argument" right off the bat, but rather to see where they mess up the basic science. The reason I use that rule is because something like 50 years of watching ID/creationists purporting to tell us anything about science tells me they always get the basic science wrong.

So the minute I see something suggesting that all interactions are equally in play no matter how far away, I know immediately the paper is crap.

It is a broad, general rule in the physics of condensed matter that distant particle interactions are "screened" by the particle interactions in the immediate vicinity of a site. That is the nature of the electromagnetic interactions in condensed matter; the largest effects take place locally. Distant charges will see a given site "cloaked" by the charges in the immediate vicinity of the site.

Single particle electromagnetic fields go as 1/r². Dipole fields fall off as 1/r³ and dipole-dipole interactions go as 1/r⁶. As more atoms become involved (multipole configurations), the interactions with more distant complexes drops off much more dramatically.

The effect is to smooth the edges of mutually interacting potential wells. You don't find spiky deep wells or deep square wells in condensed matter physics.

Many computer programs that are used to compute the configurations of clusters of atoms and molecules rely on this fact. Mean-field approximations take into account nearest neighbor interactions and then adjust more distant interactions by means of "relaxation" methods or genetic algorithms that find the self-consistent solutions that minimize energy.

Mutations at any level of condensed matter interactions are with nearest-neighbors or are bonding sites becoming lowered by some interaction with the environment that allows other molecular configurations to come into existence as long as they are relatively stable over the length of time the entire system exists. Highly energetic interactions with the environment tend to destroy the structure.

It is helpful to remember that the organic "soft matter" of living organisms is made of molecular assemblies that exist near their melting points. Their binding energies are comparable to the thermal kinetic energies of the atoms and molecules that make up the system. Any tighter binding and nothing can change over time. Any looser binding and the systems come apart.

The chemical bonds among the basic A G C T building block molecules are much larger than the other interactions involved in the folding and bonding these chains have with sites along themselves and with the aqueous environments in which they exist. It takes a bigger "whack" to change those chemical bonds than it does to change bonds due to Van der Waals types of interactions.

ID/creationist "mathematics" is more like numerology than like the applied mathematics that actually folds it the chemistry and physics. No chemists and physicists calculate like ID/creationists do.

https://www.google.com/accounts/o8/id?id=AItOawmKClPHRhWOQyCVGlIJhJOKy_uJujWHa74 · 3 October 2014

I had to read not only the post but several of Dr. Felsenstein's comments before I finally (I think) understood the basis of Dembski's argument, and why (perhaps) the University of Chicago agreed to host it. The "white noise" landscape seems pathological compared to what we observe, but to a mathematician's trained intuition, maybe not. It reminds me of the fact that the infinity of irrational numbers is larger than that of rational numbers, and that the infinity of infinitely discontinuous functions (i.e., white noise functions) is larger than that of continuous (smooth) functions.

So I see Dembski's argument as a variation of the "fine-tuning" argument: out of all possible and mostly nonsensical random sets of laws of physics, why does our universe have a smooth set? The Anthropic Principle (as mentioned by previous commenters) is one answer, which seems good enough to me. If we are to image infinite possibilities for physical laws, why not imagine these possibilities are all realized in different universes and of course we involved in one whose "because physics" allows for evolution. (As long as we speculating about stuff we don't know and probably never will).

The basic answer though is that a "designer" is not the only conceivable answer to the question, and in fact is not a good answer because an incomprehensible entity of unknown origin and unknown means of operation has no real explanatory value (absent any probative evidence of its existence).

JimV

TomS · 3 October 2014

https://www.google.com/accounts/o8/id?id=AItOawmKClPHRhWOQyCVGlIJhJOKy_uJujWHa74 said: The basic answer though is that a "designer" is not the only conceivable answer to the question, and in fact is not a good answer because an incomprehensible entity of unknown origin and unknown means of operation has no real explanatory value (absent any probative evidence of its existence). JimV

What strikes me about any of the ID arguments is in that last sentence, but I would put it more strongly: An entity of unknown operation has no explanatory value. Whatever anybody proves about the impossibility of evolution is of no benefit for ID as long as ID does not attempt to offer a solution which is not subject to the same proof. How is it that ID excludes the landscapes which (supposedly) cause a problem for naturalistic evolution? What do we know about the operation of ID which excludes such landscapes? (Can't an ID turn out such a landscape? Why not?) Without excluding some results no result is accounted for.

callahanpb · 3 October 2014

https://www.google.com/accounts/o8/id?id=AItOawmKClPHRhWOQyCVGlIJhJOKy_uJujWHa74 said: So I see Dembski's argument as a variation of the "fine-tuning" argument: out of all possible and mostly nonsensical random sets of laws of physics, why does our universe have a smooth set?

The question in the most general sense of why things are the way they are and not some other way strikes me as a philosophical question beyond science. Note that some people might argue that I'm wrong about even that much. For one thing, once a few things are understood about the universe, everything else may turn out to be a necessary consequence. Theorists are also guided by simplicity, so if a particularly elegant mathematical model matches the universe, there is a sense in which people might believe that this elegance implies a philosophical necessity. I don't really want to argue about this either way. But what amazes me is Dembski's willingness to throw away basic assumptions dating at least from Pythagoras. What I mean is that Pythagoras saw smooth, orderly geometry and rational numbers, and concluded that this was the nature of reality. To him it was scandalous (as I understand it) that the diagonal of a square could be an irrational number. This matches intuition because we observe a universe with countable quantities and ratios before considering if anything cannot be expressed as a ratio of integers. We see smooth surfaces, and many examples of symmetry such as circles. We see right angles everywhere (up and down is perpendicular to back and forth). So the existence of a well-behaved universe is among our most fundamental assumptions. It is a major concession for Dembski to admit that evolution can happen in such a universe. I agree that the Anthropic Principle would probably give us a well-behaved universe, but to be honest, I feel the necessity of order and simplicity at a more basic level. Most mathematicians are guided by aesthetic principles: while the best proof of some theorem is not guaranteed to be beautiful, many of the best ones really are, so the expectation is set up. Physicists are likewise guided by the idea that a simple model conceived out of a desire for parsimony and symmetry will match reality (which can be verified empirically) though again there may be no obvious reason to think so except for the past success of this approach. My belief in an orderly universe is just that, and nothing more. It is not belief in a designer who made it orderly. The designer argument made some sense in Paley's day when it was indeed hard to conceive of an unintentional origin of complex biological systems. But nobody even then required a designer to explanation smoothness, continuity, and symmetry. These were understood then as they are today as the expected properties of nature. So if Dembski is saying that evolution basically works assuming nature functions as people generally assume that it works, it is very hard to leverage such a view into proof of a designer.

Rolf · 4 October 2014

callahanpb said:
https://www.google.com/accounts/o8/id?id=AItOawmKClPHRhWOQyCVGlIJhJOKy_uJujWHa74 said: So I see Dembski's argument as a variation of the "fine-tuning" argument: out of all possible and mostly nonsensical random sets of laws of physics, why does our universe have a smooth set?
The question in the most general sense of why things are the way they are and not some other way strikes me as a philosophical question beyond science. [snip]

From my philosophical point of view, things are the way they are because that's the way they are. I suspect Dembski may be somewhat lacking in education with respect to physics and the ubiquitous - (and elusive?) laws of physics. In "A Different Universe", theoretical physicist, Nobel price winner Robert B. Laughlin writes:

The idea of certainty emerging through organization is deeply embedded in the culture of modern biology, and is one of the reasons my colleagues in the life sciences are so eager to declare their tolerance of uncertainty. It shows they know the scoop. What they actually mean by such statements is that microscopic uncertaity does not matter, because organization will create certainty later on at a higher level.

At the bottom, everything is physics and without a proper approach to what physiscs means for our understanding we are bound to get things wrong, like I believe Dembski does. To begin with, I don't think his "fitness landscape" really matches the realities of life in action. How many dimensions are there in his model? It seems to me that his model doesn't match real life as we know it. One might say the model was designed to get the desired results.

Joe Felsenstein · 4 October 2014

Rolf said: [Interesting discussion of biology and physics snipped. JF] To begin with, I don't think his "fitness landscape" really matches the realities of life in action. How many dimensions are there in his model? It seems to me that his model doesn't match real life as we know it. One might say the model was designed to get the desired results.

When you have as few as 1000 bases of haploid genome, and each has 4 possibilities, there are in some sense 1000 dimensions in the genotype space. I say "in some sense" because the discrete possibilities are not scales -- if you consider frequencies of the four bases at a site in a population, there are 3 dimensions per site, so 3000 dimensions overall. If you consider frequencies of all possible sequences of 1000 bases you instead have 4-to-the-1000 minus 1 dimensions. That is true for Dembski and Marks's argument and it is true for arguments I might make too. But if you take (say) 4 stretches of 250 bases each, and have each of those stretches code for a protein, then all those mutations might move you up or down on 4 scales of enzyme activity. The phenotype would then have many fewer dimensions than 1000 or 3000. However if you also consider all reasonable time courses of gene action, the number of dimensions would go up again, though limited by any smoothness constraints on these functions of time. So as you bring in physical or biochemical reality, the number of dimensions, and the extent of possible interactions of changes in the genome, shrinks a lot.

Scott F · 4 October 2014

I'm no physicist, and I did really poorly at statistics. So the way I imagine these "spiky" landscapes versus "smooth" landscapes is from a more intuitive point of view.

I envision a "spiky" landscape as a two-dimensional graph, with the third dimension being the quantity being measured. Sure, in two dimensions if you move left or right in the landscape, you might "improve" your fitness (go up to a spike), but then you're stuck. You can't go left or right, forward or back any more with "reducing" fitness, and so you'll never get to the next spike.

But let's take it down one dimension, and imagine that ours is a model of part of an actual physical landscape. Say you have only the "east-west" dimension, and your "fitness" is the vertical dimension. What you have is a line on a chart. You can go left or right, but once you've reached a local peak, you can't go anywhere else without reducing fitness. You're stuck.

But think about what this mountain-range-like 2-dimensional line represents. It represents just one slice through an actual 3-dimensional landscape. The "spike" on the 2-D plot, just happens to be one point on a local "ridge" on a 3-D plot. You may have reached a plateau in the "east-west" dimension, but you can still move in the "north-south" dimension without a reduction in "fitness". If you move in the "north-south" direction, you can reach another, different 2-D slice where you can again move in a positive direction in the "east-west" dimension.

This "model" of an actual landscape has another visceral appeal. Everyone "knows" what an actual "landscape" looks like. In reality, there aren't that many actual "peaks". There are gradual slopes and ridges. A real "landscape" doesn't look like a bed of nails. Except perhaps in some of the deserts of the American southwest, or in the Guilin region of China. But the fact that those landscapes are so exceptional, only emphasizes the smoother nature of most landscapes.

Carry that further, into the thousands of dimensions that Joe describes, and there doesn't seem to be any fundamental reason why "fitness" can't continue to improve. The likelihood of having reached a local maximum in all 1,000 dimensions at the same time seems extremely unlikely. (Perhaps the likelihood of reaching a local maximum in 1,000 dimensions can even be calculated.)

So, it seems to this untutored eye that adding more dimensions smooths out the "peaks" in the "fitness" landscape considerably. Perhaps this is one of things that is meant by many of the other posters about "reality" being a far smoother "fitness" landscape.

Scott F · 4 October 2014

In fact, one might carry that analogy to a real landscape even further. What is a "fitness" "landscape" measuring, after all? It is how "fit" the current genome is in the current environment. But just like plate tectonics, the current "fitness" landscape itself is constantly changing. Were the creature to simply stand still, and the genome not change at all, over time the current fitness "peak" would no longer be a peak, but would now be lower than the surrounding landscape. Adapt or die.

So, to model actual reality, the model must include fitness values that constantly change over time.

Then even Dembski's "spiky" model of a landscape does not preclude evolution from moving from one "peak" to another, simply because the "peaks" themselves are not constant, and change over time.

But I presume that's what ya'll have been saying all along.

Mike Elzinga · 4 October 2014

Scott F said: In fact, one might carry that analogy to a real landscape even further. What is a "fitness" "landscape" measuring, after all? It is how "fit" the current genome is in the current environment. But just like plate tectonics, the current "fitness" landscape itself is constantly changing. Were the creature to simply stand still, and the genome not change at all, over time the current fitness "peak" would no longer be a peak, but would now be lower than the surrounding landscape. Adapt or die. So, to model actual reality, the model must include fitness values that constantly change over time. Then even Dembski's "spiky" model of a landscape does not preclude evolution from moving from one "peak" to another, simply because the "peaks" themselves are not constant, and change over time. But I presume that's what ya'll have been saying all along.

That is exactly the case. The "landscape" of "peaks" or "wells" can drift over time as the environment changes; and it is often the case that there are many routes to get to a nearby peak (well). So if the environment in which a system or organism exists changes, that may mean the system/organism no longer sits near a peak/well. If the shift is large and sudden, then the system may be stranded and not be able to survive by drifting toward a nearby peak/well That nearby peak/well may not be the one on which the system was relatively stable before the environment changed; but a few changes in the system may make it possible for the system to now exist in a relatively stable configuration on this new peak/well. And since systems near their melting points always have a distribution of configurations, there may be one or more configurations out of that distribution that will be relatively stable in the new environmental landscape; in fact, different configurations may move to different peaks/wells and we then have "speciation." There are some analogies from condensed matter physics or chemistry that capture the essence of speciation and drift. Annealing is one. Raising the temperature may allow molecular rearrangements by nudging bonds over barriers that then allow the system to rearrange and settle into a lower energy state. In chemistry, chemical reactions can be "pushed" in different directions by changing the pressure, the temperature, or the concentrations of reactants. In a sense, biological adaptation is much like softening a condensed matter system so that it settles into a mold. With a physical system that simply softens, the entire system is retained as it rearranges to fit the mold. With a biological system, the changes required to fit a new "mold" are sorted out of the distribution of progeny of the original system; the progeny are the approximate replicas or surrogates of the original system that actually do the "settling in" to the new "mold." To say that a landscape is multidimensional means that there are many ways adjustments can be made that produce a system configuration that "better fits" the changed environmental conditions. A high dimensional landscape makes it difficult to predict what changes will occur and be stable.

harold · 4 October 2014

Eric said -

I think part of their strategy here is to make a silly argument opaque by using scientific terminology. Saying that smooth fitness surfaces are so rare among the types of surfaces that their presence must be explained by design is arcane enough that it sounds credible. After all, unless you work with this sort of mathematical calucation, how can you really refute that?

Eric is exactly correct, if by "part of their strategy" he means "ALL of their strategy". Dembski's model fails the most basic tests of relevance, and does so in painfully obvious ways. It essentially purports to model biological evolution, find a limitation of biological evolution, and then argue that a "designer" must be necessary to overcome that limitation. This is what all convoluted, verbose, and/or pseudo-mathematical ID output always claims to do. But it fails in all three goals. Rather than model evolution, it boils down to being another straw man version of evolution. Thus the "limitation" detected does not really exist. Furthermore, as I noted above, and as others note, even if the limitation had been of concern, no actual evidence for a designer is provided. The ubiquitous non sequitur reasoning that "if I say that something is wrong with evolution ID wins by default" is seen again. As always. The sole goal here is, in fact, to disguise evolution denial with science-y sounding language, so that some casual observer, especially one already biased against evolution, will lazily conclude that fancy-talking scientist types have shown that there is "something wrong with evolution". It is fair to describe this model as "bullshit", in fact, even in the most restricted sense of the term. It is an effort to pass off fallacious arguments by making them seem complicated, which is more or less exactly what "bullshit" means.

George Frederick Thomson Broadhead · 4 October 2014

I think to isolate Biological genomic processes to explain the development of new beings and species, to merely computational models, is forgetting, the real problem in question! What was first? The Hen or the egg?

To try and prove a designer by computational means, of how multi-cellular beings can mutate to Evolve, is a technicality! Prove it in reality, and not with models! Get a fish of today and get it to develop to land and then so on! No! Those are some of the beings that are already there! A Salamander is a middle process being!

Today we only see the downward road of genomics! I do not see an upward road!

You prove or disprove the Humunculus Argument by pure logic!

A mind proves more than merely multi-cellular transmutations! Biology Evolution and computational models are merely to specific and suffer from tunnel vision fallacies, isolating the bigger problems! It is myopic in proof capability! Hence quite pointless!

George Frederick Thomson Broadhead · 4 October 2014

Re-edited:

Today we only see the downward road of genomics! I do not see an upward road!

You prove the Humunculus Argument by pure logic!

A mind proves more than merely multi-cellular transmutations! Biology Evolution and computational models are merely to(o) specific and suffer from tunnel vision fallacies, isolating the bigger problems! It is myopic in proof capability! Hence quite pointless!

George Frederick Thomson Broadhead · 4 October 2014

ADDITIONALLY to prove or disprove a designer, does not prove the God of the Bible. Rather would disprove it more!

To be ATHEIST is simply a myopic person void of understanding how GENETIC ENGINEERING works...!!!

https://me.yahoo.com/a/JxVN0eQFqtmgoY7wC1cZM44ET_iAanxHQmLgYgX_Zhn8#57cad · 4 October 2014

George Frederick Thomson Broadhead said: Re-edited: I think to isolate Biological genomic processes to explain the development of new beings and species, to merely computational models, is forgetting, the real problem in question! What was first? The Hen or the egg? To try and prove a designer by computational means, of how multi-cellular beings can mutate to Evolve, is a technicality! Prove it in reality, and not with models! Get a fish of today and get it to develop to land and then so on! No! Those are some of the beings that are already there! A Salamander is a middle process being! Today we only see the downward road of genomics! I do not see an upward road! You prove the Humunculus Argument by pure logic! A mind proves more than merely multi-cellular transmutations! Biology Evolution and computational models are merely to(o) specific and suffer from tunnel vision fallacies, isolating the bigger problems! It is myopic in proof capability! Hence quite pointless!

I like the edit to equal incomprehensibility. Glen Davidson

Joe Felsenstein · 4 October 2014

As enthusiastic as Broadhead is, his argument is off-topic in this thread. I will send most arguing with him to the Wall.

TomS · 4 October 2014

George Frederick Thomson Broadhead said: ADDITIONALLY to prove or disprove a designer, does not prove the God of the Bible. Rather would disprove it more!

A designer is much a demiurge of Gnosticism, or a god of Pantheism or Deism. To which a promoter of ID would reply: "You don't understand ID." And, indeed, as long as there is no description of what an Intelligent Designer would be like, like how many of them are there, or what restriction there is on what they would do. As long as there is no description of what a design process is like, what are its inputs. Then nobody understands ID. If anybody were to take the concept seriously, then there would be a bunch of questions that would occur. The fact that nobody addresses any of those questions is a symptom that nobody takes it seriously. It would be a fool's errand for anyone outside the ID movement to try to take ID seriously, and discuss those questions with a straight face, for whatever would be the outcome, it would be dismissed with "you don't understand ID". The most that we outsiders have to go on are some analogies with the kinds of design that we are familiar with - but that's not the kind of design that the promoters have in mind.

Mike Elzinga · 4 October 2014

harold said: Dembski's model fails the most basic tests of relevance, and does so in painfully obvious ways. It essentially purports to model biological evolution, find a limitation of biological evolution, and then argue that a "designer" must be necessary to overcome that limitation. This is what all convoluted, verbose, and/or pseudo-mathematical ID output always claims to do.

Furthermore, ID/creationists don't even know how to count; even IF their calculations were relevant to anything. For example, take an arbitrary string of 16 bases, ACTAGACTCGTCAGGT. The probability of that arrangement is not 4¹⁶; as ID/creationists seem to always imply. There are several things wrong with such a calculation; the first being that the calculation ignores the physics and chemistry of binding energies. An A next to an A or a T next to a T or a C next to a C or a G next to a G do not have the same probabilities as, say a C next to a G or an A next to a T. Furthermore, the helical structures are determined by the binding energies of base pairs, and only certain pitches in the helical structure are allowed that minimize total energy. On top of all this, the structure forms secondary and tertiary level structures as the helical chain coils back on itself and finds various binding sites that work along its length. Those probabilities are determined by the relative magnitudes of the binding energies. But the most egregious, elementary calculation errors Dembski, et. al. make are those of over counting the number of distinct arrangements. Instead of using 4¹⁶ to maximize the number of arrangements of 16 things taken from an infinite pool made up of 4 types of things, they would need to divide that number by the number of permutations of each base in that chain. For example, all permutations of A produce the same structure; as do all permutations of every other base. ID/creationist "calculations" always aim to produce the most improbable outcome because their primary arguments are to make it appear that the known science can't explain such low probabilities. That shtick goes all the way back to Henry Morris. This pattern remains consistent through ALL ID/creationist "arguments;" misrepresent the science and then throw in a bedazzling heap of junk science to "prove" that science can't explain anything. Therefore Jesus.

callahanpb · 4 October 2014

I want to make sure I'm not misreading what Dembski said, because (repeating myself) it sounds like he conceded the truth of something like this:

Evolution works as evolutionary biologists believe it does, provided it occurs in a universe with the mathematical properties that physicists agree that our universe possesses.

If he just wants to put his name on that statement, I'd be happy to ignore everything else he has to say. The rest of it does not strike me as terribly relevant to science.

Joe Felsenstein · 4 October 2014

callahanpb said: I want to make sure I'm not misreading what Dembski said, because (repeating myself) it sounds like he conceded the truth of something like this:
Evolution works as evolutionary biologists believe it does, provided it occurs in a universe with the mathematical properties that physicists agree that our universe possesses.
If he just wants to put his name on that statement, I'd be happy to ignore everything else he has to say. The rest of it does not strike me as terribly relevant to science.

He doesn't bring physics into it. Instead he and Robert Marks just take all these properties (random association of fitnesses with genotypes) as the default condition. I am arguing that "because physics" they aren't the default condition. D&M would say that a Designer, rather than physics, is needed to make the fitness surfaces smooth enough for evolution to succeed. If they admit that physics would make the fitness surfaces smoother than random "white noise" surfaces, they could then fall back on the assertion that the Designer acted by choosing the laws of physics. At which point your characterisation of his argument would apply.

TomS · 4 October 2014

Joe Felsenstein said:
callahanpb said: I want to make sure I'm not misreading what Dembski said, because (repeating myself) it sounds like he conceded the truth of something like this:
Evolution works as evolutionary biologists believe it does, provided it occurs in a universe with the mathematical properties that physicists agree that our universe possesses.
If he just wants to put his name on that statement, I'd be happy to ignore everything else he has to say. The rest of it does not strike me as terribly relevant to science.
He doesn't bring physics into it. Instead he and Robert Marks just take all these properties (random association of fitnesses with genotypes) as the default condition. I am arguing that "because physics" they aren't the default condition. D&M would say that a Designer, rather than physics, is needed to make the fitness surfaces smooth enough for evolution to succeed. If they admit that physics would make the fitness surfaces smoother than random "white noise" surfaces, they could then fall back on the assertion that the Designer acted by choosing the laws of physics. At which point your characterisation of his argument would apply.

Let us assume, for the sake of argument, that they have proved that designers are a necessary condition. (For anything, for the smoothness of the surfaces, for the laws of physics, ...) Have they even mentioned the issue of whether designers are a sufficient condition? (For example, all the designs that we have any experience with need a producer to implement the designs and material with appropriate properties for the designs to work. And there has to be a prior state in need of design - "necessity is the mother of design [invention]".) Have they described the properties of the designers well enough to prove anything about them - even their necessity, let alone their sufficiency. Indeed, have they proved that there can be no other source of smoothness (laws, whatever)? And, have they accepted the common descent by modification of most forms of life on Earth over billions of years as the only way that intelligent designers could produce the variety that we observe? Or are they open to the idea that there may be some other way of accounting for that, and does their proof still work for other scenarios?

Joe Felsenstein · 4 October 2014

Let me clarify something I have been saying. I said a number of times that Dembski and Marks say that if evolution is to succeed, a Designer must have put information in, by choosing the fitness surface out of all the possibie fitness surfaces, most of which are white-noise fitness surfaces. In the Dembski-Marks papers they don't actually mention either a Designer or evolution -- they are talking about the need for their "active information" to be built in, for a search to succeed. It would be possible to accuse me of misrepresenting these papers if those were the only issues raised in Dembski's Chicago talk. But in the Chicago talk, Dembski is explicitly trying to relate their argument to evolution, and explicitly presenting his argument as an alternative to other Intelligent Design arguments. He has a section on "evolutionary search" and the subsequent discussion of Dawkins's Weasel example takes up much of the talk. And at the end of the session Dembski's mentor Leo Kadanoff says in conclusion:

I think the ball is in the court of people who believe in evolution. They have to deal with these questions.

which would be difficult to explain if the argument was just a technical one about general search algorithms.

Scott F · 4 October 2014

Joe Felsenstein said: Let me clarify something I have been saying. I said a number of times that Dembski and Marks say that if evolution is to succeed, a Designer must have put information in, by choosing the fitness surface out of all the possibie fitness surfaces, most of which are white-noise fitness surfaces.

Do they say what the "fitness surface" is for; what they are trying to model? Is this the "fitness" of the universe to enable evolution to happen at all, or is this the "fitness" of species (or life) to be able to evolve? If the latter, then "choosing" a fitness surface doesn't make a lot of sense, since the "fitness" of life forms is variable and continuously varying over time as the environment itself is changing. If the former, it's just the anthropic principle. Isn't it? Otherwise, they would have to be able to calculate the relative "fitnesses" of various possible universes with different possible physics or physical constants. But when you say, "if evolution is to succeed," it sounds like they are talking about the "fitness" of the universe itself. Or perhaps it's a different kind of "fitness" altogether?

Joe Felsenstein · 4 October 2014

Scott F said: Do they say what the "fitness surface" is for; what they are trying to model? Is this the "fitness" of the universe to enable evolution to happen at all, or is this the "fitness" of species (or life) to be able to evolve? If the latter, then "choosing" a fitness surface doesn't make a lot of sense, since the "fitness" of life forms is variable and continuously varying over time as the environment itself is changing. If the former, it's just the anthropic principle. Isn't it? Otherwise, they would have to be able to calculate the relative "fitnesses" of various possible universes with different possible physics or physical constants. But when you say, "if evolution is to succeed," it sounds like they are talking about the "fitness" of the universe itself. Or perhaps it's a different kind of "fitness" altogether?

It is. It's fitness as used in population genetics. Expected number of newborn offspring in the next generation, calculated for a newborn individual of that genotype. It's a function of genotype, not of which universe you are in. It's for a simple model of genotypes having constant fitnesses. Of course reality is more complex, but you need simple models to mathematically analyze what evolutionary forces will do. If there is a problem for evolution even in such a simple model then you worry that it also might be found in more realistic models. In this sense I am completely sympathetic with the use of these simple models. I have used them many times myself.

Mike Elzinga · 5 October 2014

Joe Felsenstein said: It's for a simple model of genotypes having constant fitnesses. Of course reality is more complex, but you need simple models to mathematically analyze what evolutionary forces will do. If there is a problem for evolution even in such a simple model then you worry that it also might be found in more realistic models. In this sense I am completely sympathetic with the use of these simple models. I have used them many times myself.

There is a vast difference between simple models used by scientists and the simplistic models used by ID/creationists. Biologists, for example can concoct simple models for what is actually observed in the physical universe; it is not absolutely necessary to delve into the physics and chemistry at the base of what is observed. If it takes place in the real world, then there is an ultimate physical basis behind it. In the modeling of complex systems, the phenomenological behaviors of the complex system can have a set of higher-level rules based on the physics and chemistry; they are the kind of rules that emerge out of the complexity of the system interacting with its environment. You just need to find a good model that replicates the essence of what you observe. In the case of ID/creationists, their simplistic models are all based on their own preconceptions about what they want the world to be; regardless of what is observed. When their "models" predict that something in the real world cannot happen, then they declare that science can't explain what is happening. It's a bizarre form of "logic" they use. They don't know the science, they don't observe anything, they don't know what others have observed, yet they declare their "models" tell them what cannot happen.

TomS · 5 October 2014

Mike Elzinga said: In the case of ID/creationists, their simplistic models are all based on their own preconceptions about what they want the world to be; regardless of what is observed. When their "models" predict that something in the real world cannot happen, then they declare that science can't explain what is happening. It's a bizarre form of "logic" they use. They don't know the science, they don't observe anything, they don't know what others have observed, yet they declare their "models" tell them what cannot happen.

I think that a clear of example of this is the "law of conservation of information". They first introduce a term, "information", (while intended to invoke the technical used, with different meanings in different contexts, is not given an adequate definition which would work, but let's, for the sake of argument grant them that). Then, assume that it refers to an objective property of the natural world, and that it is "conserved". (Actually, that it can decrease, but not increase. But let's not dwell on that.) Then, they "observe" that "information" can increase in the natural world. Rather than this violation of the "law of conservation of information" leading to rejection of the "law", they conclude that there must be something beyond nature which allows the violation. That something they identify (somehow) with "intelligent designers" (whatever they may be like, but let's not obscure this point by going into that). This seems to be a clear example (as I understand your point) of creating a model of the natural world, and, when the natural world does not conform to the model, it is the natural world, rather than the model, which is lacking.

harold · 5 October 2014

Mike Elzinga said:
harold said: Dembski's model fails the most basic tests of relevance, and does so in painfully obvious ways. It essentially purports to model biological evolution, find a limitation of biological evolution, and then argue that a "designer" must be necessary to overcome that limitation. This is what all convoluted, verbose, and/or pseudo-mathematical ID output always claims to do.
Furthermore, ID/creationists don't even know how to count; even IF their calculations were relevant to anything. For example, take an arbitrary string of 16 bases, ACTAGACTCGTCAGGT. The probability of that arrangement is not 4¹⁶; as ID/creationists seem to always imply. There are several things wrong with such a calculation; the first being that the calculation ignores the physics and chemistry of binding energies. An A next to an A or a T next to a T or a C next to a C or a G next to a G do not have the same probabilities as, say a C next to a G or an A next to a T. Furthermore, the helical structures are determined by the binding energies of base pairs, and only certain pitches in the helical structure are allowed that minimize total energy. On top of all this, the structure forms secondary and tertiary level structures as the helical chain coils back on itself and finds various binding sites that work along its length. Those probabilities are determined by the relative magnitudes of the binding energies. But the most egregious, elementary calculation errors Dembski, et. al. make are those of over counting the number of distinct arrangements. Instead of using 4¹⁶ to maximize the number of arrangements of 16 things taken from an infinite pool made up of 4 types of things, they would need to divide that number by the number of permutations of each base in that chain. For example, all permutations of A produce the same structure; as do all permutations of every other base. ID/creationist "calculations" always aim to produce the most improbable outcome because their primary arguments are to make it appear that the known science can't explain such low probabilities. That shtick goes all the way back to Henry Morris. This pattern remains consistent through ALL ID/creationist "arguments;" misrepresent the science and then throw in a bedazzling heap of junk science to "prove" that science can't explain anything. Therefore Jesus.

I couldn't agree more about the ID/creationist massacring of probability, but, for the sake of clarity, I will point out that if you have an urn with an effectively infinite number of balls of four different colors, in equal proportion, and you sample 16, the a priori odds of getting any exact ordered string are in fact 4^16. You divide by the number of permutations if the order doesn't matter. To determine the probability of ending up with, say, five reds, two blues, three greens, and six yellows, regardless of order, you have to divide by permutations. For an exact ordered string, the odds are 4^16. And they are the same for any exact string. The odds of getting four reds, four blues, four greens, and four yellows are much better than the odds of getting sixteen reds. But the odds or getting exactly sixteen reds are the same as the odds of getting exactly RBGYRBGYRBGYRBGY, in that order. I realize Mike knows this, of course, but it may be usefully clarifying for others. If drawing a certain color biases the probabilities of what the next color would be, that could make it more complicated. Whether that applies to spontaneous formation of nucleotide strings in some hypothetical totally enzyme free mixture of nucleotide bases at some given concentration, temperature, and pressure, I don't know. That issue just hasn't come up much in anything I've ever done because biologically relevant nucleotide strings are formed in the context of a template and enzymes like DNA or RNA polymerase. The basic problem with the simpleton creationist argument - and we all know it is fair to say that they make this argument - that "the odds of getting the human genome by random sampling are 4 to the three billion" - is that it is insanely irrelevant. Four to the three billionth power is actually the correct probability of generating an exact sequence of three billion nucleotides by choosing each individual base, sequentially, by random sampling from an infinite pool of equally concentrations of all bases for each position, but that is utterly irrelevant to any situation in biomedical science because no-one ever suggested that such a contrived scenario ever occurs. It's actually more of a stupid argument against abiogenesis, but no model of abiogenesis ever proposes instantaneous appearance of a modern genome by random sampling for each position, either.

harold · 5 October 2014

In the Dembski-Marks papers they donât actually mention either a Designer or evolution â they are talking about the need for their "active information" to be built in, for a search to succeed. It would be possible to accuse me of misrepresenting these papers if those were the only issues raised in Dembskiâs Chicago talk. But in the Chicago talk, Dembski is explicitly trying to relate their argument to evolution, and explicitly presenting his argument as an alternative to other Intelligent Design arguments. He has a section on "evolutionary search" and the subsequent discussion of Dawkinsâs Weasel example takes up much of the talk.

I got that, but I did fail to note something interesting. This shows an exponential decay in the number of remotely positive claims that certain ID/creationists make, or, equivalently, a logistic increase in the weaseling and reliance on code and implication of their arguments. "Creation science" asserted that science actually supports a 6000 year old Earth and Noahic flood, with the actions of a right wing American Protestant God (not to be confused with the liberal American Protestant God of, say, Dr. Martin Luther King) as the only possible explanation. Then it decayed rather rapidly to standard ID - which always claims that evolution can't explain some change in a biological lineage (always falsely) and always makes a non sequitur jump to an anonymous and uncharacterized "designer" who is understood by all to secretly be the right wing American Protestant God mentioned above. Now we see a more gradual decay to a claim that maybe the changes in biological lineages can be explained scientifically, but something must be secretly diddling with the environment in the background. That something being understood by all reasonable and informed observers to be, of course, "the designer", and therefore, You Know Who. But direct claims about molecular genetic evolution are more disguised, and to a lesser degree even the word "designer" is now implied rather than openly stated. It's still the same crap. Analogy - Cookies are missing from the jar, Bob has cookie crumbs on his face and fingers and is belching, and a reliable videotape shows Bob taking cookies from the jar. The Cookie Monster is nowhere to be seen. Creation Science version - We know in advance that cookies are only stolen by the Cookie Monster. Therefore by definition there is something wrong with the video, there is something wrong with the crumb evidence, and when "interpreted differently" all evidence proves that the Cookie Monster took the cookies. Vintage 1998 ID version - Despite the tape and crumbs, it is somehow theoretically impossible for Bob to take cookies, so we can ignore the tape and crumbs. Something else must have taken the cookies, and we'll just call it the "cookie taking designer", but hey, if you think that sounds like the Cookie Monster, more power to you. 2014 Dembski version - Okay, maybe Bob himself can theoretically take cookies, but there's something mysterious about the jar that makes it impossible for him to take these cookies. The cookies are gone, though, I'm still saying Bob couldn't have taken them damn the evidence, and hey, I may or may not even always openly say the words "cookie taking designer". But if you still think it sounds like the Cookie Monster, let's share a chuckle.

Joe Felsenstein · 5 October 2014

A few corrections: TomS and harold have probabilities and odds with values like 4-to-the-16th. Hey guys, probabilities can't be greater than 1. I'm sure you meant to say 1 in 4-to-the-16th or odds of 4-to-the-16 against. If you find any event with probability greater than 1, let me know: I want to place a few bets. I wouldn't characterize Dembski's argument in the cookie-jar case the way harold has. Rather:

Okay, even if Bob himself actually took the cookies, this occurred because the shape and color of the jar was so tempting that it made him do that. And the jar was designed by the Cookie Monster to be tempting.

Dembski's and Marks's most recent argument does not say that evolution by natural selection and other evolutionary forces cannot happen. It says that even if it happens, the fitness surface contains information that must have been put there by a Designer. (And, no I am not accusing Robert Marks of making off with any cookies. I am sure he wouldn't do that.)

harold · 5 October 2014

If drawing a certain color biases the probabilities of what the next color would be, that could make it more complicated.

To expand very briefly on this, it would change the distribution of the exact strings of sixteen. Instead of each being equally likely with an a priori probability of 4^16, some would be much more likely than this, and others much less likely. The probability of an individual string would still be determined by what amounts to a multinomial process, just a more complicated one. The expected probability of any totally arbitrarily chosen string would be 4^16, but now, some arbitrarily chosen strings would be more likely than that, and some less. However, the real problem with creationist "odds against generating a contemporary biological sequence of amino acids or nucleic acids by pure random sampling" is not so much this nuance, but rather, that the idea that biological amino acid or nucleic acid sequences are determined by random sampling, one at a time, from some implied pool of equally concentrated amino or nucleic acid bases, is a straw man absurdity. That is categorically not how such sequences have ever formed in living cells. Nor does any model of abiogenesis suggest that proto-cells formed by pure random sampling of components from an imaginary hat or urn, one molecule at a time. (I will note that during the replication of a strand of DNA, the probability of some individual mutation occurring is extremely well modeled as a random variable. That's absolutely true, and the fact that some mutations have greater frequency than others is irrelevant - drawing a ball from an urn of ten thousand white balls, one hundred green balls, and one black ball is still an example of random variable. The mutations that can occur are obviously constrained by the original DNA sequence, of course, just as which ball you can draw is constrained by which type of balls are in the urn.)

harold · 5 October 2014

TomS and harold have probabilities and odds with values like 4-to-the-16th. Hey guys, probabilities can't be greater than 1. I'm sure you meant to say 1 in 4-to-the-16th or odds of 4-to-the-16 against. If you find any event with probability greater than 1, let me know: I want to place a few bets.

Yes, that is of course what I meant. I made the same typo above, as well. However, I stand by the general content of those comments. I was thinking "four to the negative sixteen" of course, but didn't make a negative sign.

I wouldn't characterize Dembski's argument in the cookie-jar case the way harold has. Rather: Okay, even if Bob himself actually took the cookies, this occurred because the shape and color of the jar was so tempting that it made him do that. And the jar was designed by the Cookie Monster to be tempting.

I agree with this. I think my analogy was imperfect. Dembski's latest thing is more like "Okay, okay, so Bob could take the cookies. Therefore I claim that the jar had to be magically modified for him to be able to take the cookies. I'll stop talking now, but I think we all know who I'm implying must have modified that jar..." I don't mean to say that your version isn't equally correct. Basically he's saying "I won't say this out loud but it seems as if our old arguments about living organisms being irreducibly complex and whatnot have been shot to Hades, so now I'll switch it up and say that the environment is what needs to be magically modified.

harold · 5 October 2014

Dembski may have gotten himself in trouble with his own team again, if any of them bother to listen.

To continue the analogy, they aren't satisfied with vague claims that the Cookie Monster exists even though Bob took the cookies. That's a bunch of "Cookie Monster evolutionist" stuff for the liberals at Unitarian Universalist seminaries. They want denial that Bob took the cookies.

Joe Felsenstein · 5 October 2014

And ta-da! Here comes the ID reaction!

Over at Uncommon Descent their regular commenter "News" informs us of PZ Myers' reaction at Pharyngula. PZ described the video of Dembski's lecture as too tiresome for him to concentrate on, so he pointed to this post as being a detailed refutation of it. And he cited parts of my PT post (the original post for this thread) and not just a few words, but 426 words of it in all.

But the writer of News, who is almost certainly Denyse O'Leary, does not mention that Myers' Pharyngula post cited any argument against Dembski's. This brilliant piece of science journalism manages to leave the impression that all Myers did was use his boredom as his only argument against Dembski.

It then goes on to ascribe the objections to ID to "Darwin's followers", including "union high school teachers regurg[itat]ing talking points" and lawyers, who "ready themselves to enforce Darwinism in court and inflict it on tax-funded compulsory school systems."

As a grandson of a union man, and son of a union man and a union woman, and myself a former secretary of an American Federation of Teachers local, I should take offense at O'Leary's remarks. But I'm above all that ... well ... no, dammit, I'm not. I do take offense.

We may expect a weightier reaction from the ID folks later.

harold · 5 October 2014

Joe Felsenstein said: And ta-da! Here comes the ID reaction! Over at Uncommon Descent their regular commenter "News" informs us of PZ Myers' reaction at Pharyngula. PZ described the video of Dembski's lecture as too tiresome for him to concentrate on, so he pointed to this post as being a detailed refutation of it. And he cited parts of my PT post (the original post for this thread) and not just a few words, but 426 words of it in all. But the writer of News, who is almost certainly Denyse O'Leary, does not mention that Myers' Pharyngula post cited any argument against Dembski's. This brilliant piece of science journalism manages to leave the impression that all Myers did was use his boredom as his only argument against Dembski. It then goes on to ascribe the objections to ID to "Darwin's followers", including "union high school teachers regurg[itat]ing talking points" and lawyers, who "ready themselves to enforce Darwinism in court and inflict it on tax-funded compulsory school systems." As a grandson of a union man, and son of a union man and a union woman, and myself a former secretary of an American Federation of Teachers local, I should take offense at O'Leary's remarks. But I'm above all that ... well ... no, dammit, I'm not. I do take offense. We may expect a weightier reaction from the ID folks later.

Have I ever mentioned that ID/creationism is massively associated with a certain political ideology? I feel as if I may have mentioned that before.

Joe Felsenstein · 5 October 2014

And for comic relief, in the comments at Denyse O'Leary's "News" thread (the one mentioned above) "Joe" says that

Joe Felsenstein chides Dembski for a lack of evidence yet Joe has never presented any evidence for natural selection actually doing something. The point is no one can refute Dembski without providing that evidence. PZ cannot offer up any evidence tat refutes Dembski, thatâs a certainty

("Joe" is always setting himself up as the arbiter and demanding that we show evolution "doing something" where he gets to define what "doing something" means). "Joe" is also reading-challenged. For never, anywhere in this thread, did I "chide Dembski for a lack of evidence". "Joe" just made that up. A more substantive reaction at the UD thread was user vjtorley, who did note that PZMyers had based himself on my argument, and vjtorley provided a link to the thread here. Torley is one of the more serious commenters at UD and he seems to make an honest attempt to engage with arguments and discuss their implications. Torley in effect corrected Denyse O'Leary but he did not make any argument supporting Dembski's.

TomS · 5 October 2014

Joe Felsenstein said: A few corrections: TomS and harold have probabilities and odds with values like 4-to-the-16th. Hey guys, probabilities can't be greater than 1. I'm sure you meant to say 1 in 4-to-the-16th or odds of 4-to-the-16 against. If you find any event with probability greater than 1, let me know: I want to place a few bets.

[blush] As far as making any bets when the probability is greater than 1, you should be aware of the payoff. Just as you can expect 0 (minus the vigorish) for winning on a sure thing, the return due on a greater-than-one bet is negative. I don't know what it would be for something with a negative probability (maybe an imaginary number). Thank you.

Mike Elzinga · 5 October 2014

harold said:
TomS and harold have probabilities and odds with values like 4-to-the-16th. Hey guys, probabilities can't be greater than 1. I'm sure you meant to say 1 in 4-to-the-16th or odds of 4-to-the-16 against. If you find any event with probability greater than 1, let me know: I want to place a few bets.
Yes, that is of course what I meant. I made the same typo above, as well. However, I stand by the general content of those comments. I was thinking "four to the negative sixteen" of course, but didn't make a negative sign.

Just to clarify, I made the same typo myself. I neglected to state the probability as 1/4¹⁶. I was concentrating on the damned tags and missed the one-divided-by. I hope that was clear from the way I used it in the rest of my comment in which I was referring to numbers of possibilities. As you state, ID/creationists are counting exact sequences taken from an infinite pool made up of several types; whether they be Scrabble letters or die or coins. That characterization in and of itself is a grotesque misrepresentation of how atoms and molecules come together. They don't come marching one-by-one out of an urn and settle in next to the one that came before. But again, this mischaracterization of the basic physics and chemistry of atoms and molecules springs from their Fundamental Misconceptions of the laws of thermodynamics, which they inherited from Henry Morris and is consistent with their doctrine of the degradation of the universe since the Fall of Man. In the real world of chemistry and physics and statistical mechanics, one divides the number of permutations of things by the numbers of permutations of each of the things that appear more than once. For example, N!/(n₁! n₂! n₃! ... n_k!), where n₁ + n₂ + ... + n_k + ... = N. Energy states or assemblies don't come out of an urn in the real world of chemistry and physics.

Mike Elzinga · 5 October 2014

Arg! this editor is getting to be a pain.

harold · 5 October 2014

I realized I made another typo...

I couldnât agree more about the ID/creationist massacring of probability, but, for the sake of clarity, I will point out that if you have an urn with an effectively infinite number of balls of four different colors, in equal proportion, and you sample 16, the a priori odds of getting any exact ordered string are in fact 4^16. You divide by the number of permutations if the order doesnât matter. To determine the probability of ending up with, say, five reds, two blues, three greens, and six yellows, regardless of order, you have to divide by permutations.

You would multiply by the number of permutations. Without typos - I couldnât agree more about the ID/creationist massacring of probability, but, for the sake of clarity, I will point out that if you have an urn with an effectively infinite number of balls of four different colors, in equal proportion, and you sample 16, the a priori odds of getting any exact ordered string are in fact 4^-16. You multiply by the number of permutations if the order doesnât matter. To determine the probability of ending up with, say, five reds, two blues, three greens, and six yellows, regardless of order, you have to multiply by permutations. For an exact ordered string, the odds are 4^16. And they are the same for any exact string. The odds of getting four reds, four blues, four greens, and four yellows are much better than the odds of getting sixteen reds. But the odds or getting exactly sixteen reds are the same as the odds of getting exactly RBGYRBGYRBGYRBGY, in that order.

Mike Elzinga · 5 October 2014

Joe Felsenstein said: Dembski's and Marks's most recent argument does not say that evolution by natural selection and other evolutionary forces cannot happen. It says that even if it happens, the fitness surface contains information that must have been put there by a Designer. (And, no I am not accusing Robert Marks of making off with any cookies. I am sure he wouldn't do that.)

Yet if you look at the Dembski and Marks paper, they are talking about combination locks, and other exact sequences. "Methinks it is like a weasel" is an exact sequence, and ID/creationists sure go after that in a way that shows they don't comprehend the point of Dawkins' little program. I think in the past I have referred to ID/creationist probability calculations as being based on assemblies coming out of an ideal gas of inert objects, and a process that starts all over again once something isn't "correct." Perhaps it is worse than that; perhaps they don't think at all.

harold · 5 October 2014

As you state, ID/creationists are counting exact sequences taken from an infinite pool made up of several types; whether they be Scrabble letters or die or coins. That characterization in and of itself is a grotesque misrepresentation of how atoms and molecules come together. They donât come marching one-by-one out of an urn and settle in next to the one that came before.

We both said it, but it's worth repeating, so I'll do it by quoting you. This is not directly related to Dembski's exact model under discussion here, but a valid thing to bring up about ID/creationist "mathematical" models in general. And Dembski's model shows the same general flaw. It's not that their math is always internally incorrect, although of course it often is. But sometimes not. It's that their model is always inappropriate. It's trivially obvious that drawing amino acid identities out of a hat like scrabble letters would be an extremely hard way to generate some precise sequence of 100 amino acids, but no-one ever said any such thing ever happens.

Mike Elzinga · 5 October 2014

harold said: This is not directly related to Dembski's exact model under discussion here, but a valid thing to bring up about ID/creationist "mathematical" models in general. And Dembski's model shows the same general flaw.

Yeah; I understand the point that Joe is making about the fitness landscape. I think I have heard other ID/creationists assert that their deity put the "active information" into the universe at the beginning; i.e., front-loaded into the laws of physics. If that is the case, whether it is hidden in the landscape or hidden in the laws of physics, then it is basically a god-of-the-gaps argument and has no relevance to the conduct of real science. It may make them feel they have won an argument, but they still canât teach their sectarian doctrines in the public schools. And they sure don't seem to be able to do any science. Of course, I would still like to know how "information" pushes atoms and molecules around without being detected.

Mike Elzinga · 5 October 2014

TomS said: This seems to be a clear example (as I understand your point) of creating a model of the natural world, and, when the natural world does not conform to the model, it is the natural world, rather than the model, which is lacking.

Yes; that is pretty much it. Since the formal founding of the Institute for Creation "Research" by Henry Morris back in the 1970s, "scientific" creationists and their spin-offs, the ID/creationists, have been trying to give their sectarian dogma the pizzazz of science. The original idea was that if they could make their sectarian dogma look like science, they could teach it in the public school science curriculum. The result, after something like 50 years of battling in the courts, has been the construction of a Potemkin village made up of cargo-cult science, complete with imitation peer-reviewed journals and "proper academic citations." They seem to think that the appearance of science can be marketed as real science; and that kind of thinking seems to have worked in the gullible communities of fundamentalists. There is now an entire cottage industry - e.g., AiG, the ICR and the DI - that panders to this crowd as well as the Right Wing.

DS · 5 October 2014

Joe Felsenstein said:
DS said: ... I understand that it is a theoretical argument. However, if it does not apply to the real world, it is worthless. It is not adequate to envision some circumstances under which evolution might not work. It is necessary to demonstrate that it cannot work under realistic conditions. If one never measures any fitness values or has any idea what he real fitness landscape looks like, then it seems counter productive to try to infer that evolution cannot happen. And it doesn't matter if the conditions required for evolution to work are rare if they actually exist. That's just another bad application of the anthropic principle.
Dembski and Marks are not saying evolution can't happen. They are saying that even if it can, a Designer had to carry out a Design Intervention to set up the fitness surface so that it is the right shape for evolution to succeed. They're wrong "because phyiscs", but that's what they're saying.

Well in that case they must admit that evolution did happen and that it happened because god wanted it to happen. So I guess they shot themselves in the foot again.

callahanpb · 5 October 2014

Joe Felsenstein said: It says that even if it happens, the fitness surface contains information that must have been put there by a Designer.

So I guess he is not conceding as much as I thought. There is a difference between saying that a fitness surface needs to be well-behaved for any process such as evolution to function, and saying that the surface itself is designed to lead to particular kinds of results. It sounds like his "information smuggling" objections all over again.

Mike Elzinga · 5 October 2014

Mike Elzinga said:
TomS said: This seems to be a clear example (as I understand your point) of creating a model of the natural world, and, when the natural world does not conform to the model, it is the natural world, rather than the model, which is lacking.
Yes; that is pretty much it.

Just to clarify even further on that point. As Joe mentioned above:

Dembskiâs and Marksâs most recent argument does not say that evolution by natural selection and other evolutionary forces cannot happen. It says that even if it happens, the fitness surface contains information that must have been put there by a Designer.

ID/creationist misrepresentations and "calculations" are designed to convince the gullible that the science isn't up to the task of explaining what is going on. ID/creationists continue to mischaracterize the scientific concepts as they have always done. This tactic has been their shtick ever since Henry Morris started it back in the 1970s. The point is to justify and "validate" their own mistrust of science, and to do it by giving the appearance of doing "better" sectarian science that comports with sectarian beliefs. Yet none of this ID/creationist bunch of leaders actually does any real science; they just crank out books and papers kvetching about real science and real scientists.

John Harshman · 5 October 2014

Joe Felsenstein said: I said a number of times that Dembski and Marks say that if evolution is to succeed, a Designer must have put information in, by choosing the fitness surface out of all the possibie fitness surfaces, most of which are white-noise fitness surfaces.

How would a deity go about designing a fitness surface? Since the surface is a response to all environmental variables, wouldn't the designer have to specify all those variables? And since those variables change over time, wouldn't he have to guide all those changes. Pretty much, this hypothesis seems to me to require constant orchestration of everything in the world, unless of course the world is capable of running itself, which invalidates the claim.

And at the end of the session Dembski's mentor Leo Kadanoff says in conclusion:
I think the ball is in the court of people who believe in evolution. They have to deal with these questions.
which would be difficult to explain if the argument was just a technical one about general search algorithms.

Whoa. Doesn't this imply that Kadanoff is an IDiot?

Mike Elzinga · 5 October 2014

John Harshman said: Whoa. Doesn't this imply that Kadanoff is an IDiot?

That one puzzles me. Leo Kadanoff has done significant research in physics and was president of the American Physical Society in 2007. I often enjoyed his columns in Physics Today. I certainly didn't have the impression he is an IDiot. I would have figured he would see through Dembski's "math." He apparently was Dembski's thesis advisor; but I don't know how to interpret that remark about the ball being in the court of people who believe in evolution. It looks like I may have to do some more digging about Kadanoff's religious beliefs. I don't know what Kadanoff knows about the history of ID/creationist movement. A lot of physicists don't pay much attention to sectarian political shenanigans.

DS · 5 October 2014

Well if they admit that evolution can indeed occur, given a certain type of fitness landscape, how does that get them anywhere? How dies saying that that fitness landscape could never happen without divine intervention affect evolutionary theory at all? In order for e their "argument" to have any meaning, they must admit that evolution is not only is possible, but that if it occurred, it must have occurred because god wanted it to occur. Now no one cares one bit whether they believe in a god or not. The main point is that they must admit that evolution can occur or their whole "argument" falls apart. This seems to be a major concession on their part, since they seem to have been saying just the opposite for years. They should be reminded of this every time they use the "argument".

TomS · 5 October 2014

Mike Elzinga said: ID/creationists continue to mischaracterize the scientific concepts as they have always done. This tactic has been their shtick ever since Henry Morris started it back in the 1970s. The point is to justify and "validate" their own mistrust of science, and to do it by giving the appearance of doing "better" sectarian science that comports with sectarian beliefs.

Actually, this goes back to the early 19th century, and in its origins was quite straightforward. The science as described by Francis Bacon was their model, and the "common sense" philosophy of Thomas Reid. This was developed into a "science" of Christian theology in the "Princeton Theology" (in Princeton Theological Seminary, with no relation to Princeton University) by people like Archibald Alexander and Charles Hodge. As I understand it, they took the Bible to supply their data, and then applied the Baconian methodology of induction. This is one of the underpinnings of Fundamentalism. (You can find brief descriptions of this in various Widipedia articles.) Since then, philosophy of science has developed far beyond Bacon. (Although you can still hear people talking about the "scientific method" and "induction" as the hallmarks of science.) I don't know how deep today's creationists go into studying the Princeton Theology.

Mike Elzinga · 5 October 2014

TomS said: I don't know how deep today's creationists go into studying the Princeton Theology.

Some of the ID/creationists, including Dembski, give credit to A.E. Wilder-Smith. I had heard only a couple of talks on film by Wilder-Smith, well before "scientific" creationism and the ID/creationist movement became organized and political. I think Wilder-Smith was behind some of the more "updated" versions of - what does one call it - perhaps "theistic" science? I don't know what he would have thought of the current ID/creationist movement. When "scientific" creationism got started back in the 1970s, it had an overt and unapologetic hostility toward science and the science community about it. I still have some of the early writings of Morris, Gish, and Parker. It is clear they were trying to tap into some deep ideological hatred among fundamentalists toward the secular world. Sputnik had happened and a huge push to upgrade the science educational standards in the US was under way at the time. That became a political issue for the creationists. I wouldn't characterize any of the ID/creationists of being scholarly. Morris and Gish were often rather bumptious. Gish, in particular, was quite thuggish in his bullying of teachers in Kalamazoo, Michigan. They appeared to be on a mission of hate. I think some of the later spin-offs in the form of "Intelligent Design" tried to cultivate a more "scholarly" image; but the "genetic" connection to the earlier ideology and the misconceptions and misrepresentations of science by the "scientific" creationists remain to this day.

Mike Elzinga · 5 October 2014

Well, as I suspected from my prior readings of Kadanoff's writings, Kadanoff is not an IDiot. Here is piece written by him in Physics Today back in 2006. In a specific reference to the Dover trial, he writes on Page 2

Other scientific subjects suffer from complementary difficulties. For example, biology boasts a larger proportion of fully qualified teachers, but they are often shackled by rules and policies that prevent them from presenting a full picture of their science. Most of biology is closely tied to evolutionary theory, which provides a framework for integrating the specific knowledge in biology. Many states, school districts, and parents' associations discourage the teaching of modern evolutionary knowledge because it conflicts with strongly held religious beliefs. The activities of the Dover, Pennsylvania, school board have been closely examined because of a lawsuit about board strictures against the teaching of evolution. Judge John Jones's decision in that lawsuit noted, with opprobrium, that the board's interference was related to a quasi-political effort by the intelligent design movement. That movement contains a wide variety of enthusiastic believers opposed to the dissemination of knowledge about evolution. Many adherents of ID are even opposed to scientific discussions about the age of the universe. Thus, religion-based conflicts can impede a full education in science.

(Emphasis added) One has to wonder what Kadanoff now thinks of Dembski; especially given Dembski's cowardly role in that trial and in the construction of that "fart video" Dembski put up on the Internet when Judge Jones ruled against the creationist school board.

Mike Elzinga · 5 October 2014

I took a Dramamine pill and watched the video. Dembski is really befuddled.

I suspect Kadanoff is no fool in inviting Dembski to give that talk.

While I couldn't hear all the questions near the end - starting at minute 53 - I heard enough to know that several were related to Kadanoff's question about excluding possibilities. The question was essentially "What is NOT a target?" and Dembski didn't get it. Another person asked, "What if the solution set is empty?" Dembski was like a deer in the headlights.

It appears to me that the audience had Dembski's number; the questions were pointed and Dembski kept insisting that the search was for something teleological.

I've seen crackpot talks like this before; the knowledgeable audience simply monitors the crackpottery to see what is currently being run up the flagpole. Sometimes it is better to let the crackpot think he is getting away with his bull pucky.

But we have already discussed the problems with Dembski's notions about evolution here on Panda's Thumb a number of times before.

Joe Felsenstein · 6 October 2014

The audience seemed to be puzzled by Dembski's argument, but I didn't hear anyone really nail him.

The reason I think it is not sensible to consider evolution in a simple model of genotypes with constant fitnesses to be as "search" is:

1. You can make the "target" the genotype of highest fitness, or the set of genotypes tied for highest fitness. Even in a haploid case, going uphill on the surface won't necessarily get there ...

2. So then what is the target? All the tops of local peaks? Including ones that are quite low?

3. Evolution can improve fitness a lot without necessarily finding the optimal genotype and phenotype. In fact, we are probably not as optimal as we could be -- we are the result of climbing the nearest peak rather than the highest one. (And of course of moving on changing fitness surfaces).

So one can have a reasonably successful evolutionary outcome without even thinking about searching for a specific target. Simulated evolution is not necessarily an "evolutionary search".

The issues I have raised about physics predisposing to smoother-than-random fitness surfaces are separate from these issues of whether (or how) to think about evolution as a search. Dembski had some very unclear statement about evolution being teleological in that it searched for teleology. Me, I'm just searching for the meaning of that statement, and haven't found it yet.

Gordon Davisson · 6 October 2014

Joe Felsenstein said: [in reply to to ksplawn] You're right -- it is very very hard to set up any physically realizable system that has that much interaction among its parts. The exception would be installing a cryptographic system between the genotype and the phenotypes. If the (haploid) genotypes were binary strings such as 100101110 ... 010111 and then one encrypted them, and got out a dramatically different string such as 0011101101010100 ... and took the individual digits as the phenotypes of different characters, and had the fitness be some reasonable function of the phenotype values, then the fitness surface as a function of the genotype string would be of the white-noise variety.

Actually, "encryption" isn't guaranteed to do this, and for essentially the same reason: limited interaction between parts. Most modern encryption systems are based on block ciphers, which means that they take a fixed-size block of data (e.g. AES uses a 128-bit block), and stir that well. If you're encrypting more bits than that, there are a number of "modes" you can use to extend it: ECB ("electronic codebook") mode simply encrypts each 128-bit block independently, meaning that the fitness of each block would be (at least potentially) independent, and your argument would still apply. CBC (cipher block chaining) mode (and a number of similar modes), in which each block is mixed with the preceding blocks, so changes in one block affects the encryption of all later blocks. But even in this case, your argument applies because an evolutionary search could potentially optimize the first block, then the second, etc. CTR (counter) mode, in which an encrypted counter is exclusive-ORed with the bits of the plaintext... meaning that a single-bit change in the plaintext data only changes the corresponding bit in the encrypted data (yes, there are some encryption scenarios where that's ok). Evolution's efficiency would be completely unaffected by this type of encryption. Note that, in general, cryptosystems have to be carefully designed to spread around causal connections (see Shannon's principle of diffusion), so I'm finding it hard to see why one should expect an undesigned system (including undesigned laws of physics) to exhibit the sort of complete interdependence that Dembski et al assume...

Mike Elzinga · 6 October 2014

Joe Felsenstein said: The issues I have raised about physics predisposing to smoother-than-random fitness surfaces are separate from these issues of whether (or how) to think about evolution as a search. Dembski had some very unclear statement about evolution being teleological in that it searched for teleology. Me, I'm just searching for the meaning of that statement, and haven't found it yet.

I was surprised Dembski kept insisting on that as well, especially in front of Kadanoff. One of the questioners asked Dembski if he had presented his ideas to astrophysicists. Dembski seemed confused and asked what the target would be, and the questioner gave the example of stars. Then, as I recall, that got linked to the question of the ruled-out alternatives to stars; i.e., "not stars. The questioner also tried to point out that stars are a simpler system than living organisms, so why wouldn't Dembski's notions apply to stars? I think I understand the genesis of that person's question; it's the well-known fact that matter condenses to form more complex structures with new properties emerging out of the complexity. I don't think Dembski got the point. He kept insisting on portraying evolution as a search for teleological systems. As near as I can tell, he views living systems as having purpose. He doesn't see the connections of atoms and molecules condensing into complex organic structures, with new emerging properties, as being on a continuum with the formation of stars in which simpler systems of matter are built up into more complex systems that are subsequently blown out into the universe to be cycled into still more complex structures. Gravity and electromagnetic interactions play no part in Dembski's calculations; for him and his followers, it's all inert objects assembling randomly into some "purposeful order." He doesn't seem to have a clue about now physics and chemistry work despite having been mentored by Kadanoff. I find it hard to imagine that Kadanoff would not find that a bit strange.

harold · 6 October 2014

TomS said:
Mike Elzinga said: ID/creationists continue to mischaracterize the scientific concepts as they have always done. This tactic has been their shtick ever since Henry Morris started it back in the 1970s. The point is to justify and "validate" their own mistrust of science, and to do it by giving the appearance of doing "better" sectarian science that comports with sectarian beliefs.
Actually, this goes back to the early 19th century, and in its origins was quite straightforward. The science as described by Francis Bacon was their model, and the "common sense" philosophy of Thomas Reid. This was developed into a "science" of Christian theology in the "Princeton Theology" (in Princeton Theological Seminary, with no relation to Princeton University) by people like Archibald Alexander and Charles Hodge. As I understand it, they took the Bible to supply their data, and then applied the Baconian methodology of induction. This is one of the underpinnings of Fundamentalism. (You can find brief descriptions of this in various Widipedia articles.) Since then, philosophy of science has developed far beyond Bacon. (Although you can still hear people talking about the "scientific method" and "induction" as the hallmarks of science.) I don't know how deep today's creationists go into studying the Princeton Theology.

However, one must avoid anachronistic judgment of those who lived in the past. "Newton was a creationist" but almost all the data arguing against creationism had not been discovered yet in Newton's time. I do agree that the first half of the nineteenth century saw a foolish religious objection to early scientific geology. And that objection often emanated from mainstream, highly educated clerical figures, not Ken Ham types. But today those denominations are not YEC. It's true that earlier stuff may have influenced the Morris/Gish movement, of which Dembski is essentially a continuation, but it was mainly a profoundly backwards reaction to features of the sixties and seventies. 1) Sputnik, as Mike notes, and the sudden increase in the prestige of science. 2) Increased rights for women, and 3) Last but very much not least, the support of the Civil Rights Movement by mainstream churches. Now, mainstream churches had tended to be somewhat conservative, but not reliably right wing, before that. They were a mix. A reverend in an Episcopalian church with a lot of millionaires in the pews might make sneering comments implying that unions or Franklin Delano Roosevelt were sinful. A reverend in a different Episcopalian church might imply the opposite, and condemn the rich. Reactionaries were able to tolerate the ambiguity. But the near-universal support for civil rights by mainstream religion was a source of panic. There was as desperate need to regain some sort of moral rationalization. Hence the religious right was born. And Creation Science was an early and integral part of that. You have to understand that if people "interpret" the Bible, any fool can see that it's got a lot of nasty parts. But basically, only God, or someone directly commanded by God, is allowed to be nasty. The rest of the time, you're supposed to be obeying the Ten Commandments and the Golden Rule. The Bible may assume slavery, but it's desperately difficult to argue that the character Jesus would have favored American slavery over emancipation, for example. The way to make the Bible harsh and reactionary is to insist that the more brutal passages need to be "taken literally". That's what it's all about. This has always been done, but it became a panicked necessity when mainstream churches turned "liberal" on civil rights. I'm not religious and if someone thinks that my explanation of the need for "literalism" to justify harsh right wing policy above is "being too nice to religion", so be it. My goal is not to defend any religion. The simple point I'm making is that there is a reason why right wing authoritarians favor a rigid (but very selective) "literal" interpretation of the Bible.

John Harshman · 6 October 2014

Mike Elzinga said: Well, as I suspected from my prior readings of Kadanoff's writings, Kadanoff is not an IDiot. Here is piece written by him in Physics Today back in 2006. In a specific reference to the Dover trial, he writes on Page 2
Other scientific subjects suffer from complementary difficulties. For example, biology boasts a larger proportion of fully qualified teachers, but they are often shackled by rules and policies that prevent them from presenting a full picture of their science. Most of biology is closely tied to evolutionary theory, which provides a framework for integrating the specific knowledge in biology. Many states, school districts, and parents' associations discourage the teaching of modern evolutionary knowledge because it conflicts with strongly held religious beliefs. The activities of the Dover, Pennsylvania, school board have been closely examined because of a lawsuit about board strictures against the teaching of evolution. Judge John Jones's decision in that lawsuit noted, with opprobrium, that the board's interference was related to a quasi-political effort by the intelligent design movement. That movement contains a wide variety of enthusiastic believers opposed to the dissemination of knowledge about evolution. Many adherents of ID are even opposed to scientific discussions about the age of the universe. Thus, religion-based conflicts can impede a full education in science.
(Emphasis added) One has to wonder what Kadanoff now thinks of Dembski; especially given Dembski's cowardly role in that trial and in the construction of that "fart video" Dembski put up on the Internet when Judge Jones ruled against the creationist school board.

I am unable to reconcile Kadanoff's 2006 statement with his 2014 one, unless the latter was intended ironically.

Joe Felsenstein · 6 October 2014

callahanpb said: So I guess he is not conceding as much as I thought. There is a difference between saying that a fitness surface needs to be well-behaved for any process such as evolution to function, and saying that the surface itself is designed to lead to particular kinds of results. It sounds like his "information smuggling" objections all over again.

The Information Smuggling objection is a more general description which the Search For a Search (SFS or S4S) makes a little more specific, and the Conservation of Information (CoI) adds in some theorems that explain "active information" and compute how much of it there will be, which depends on the smoothness of the fitness surface. It's all basically the same argument, some details being fleshed out as one goes along. If one has a reason to reject the equiprobability of all ways of association of fitnesses with genotypes, "because physics", one can pretty mich forget about the CoI and one is back to the earlier forms of the argument.

TomS · 6 October 2014

As Harold knows, I am pretty much in agreement with him, and have nothing to cavil with in his latest

But I really don't know how much today's fundamentalists are aware of their history.

Joe Felsenstein · 6 October 2014

John Harshman said:
Mike Elzinga said: Well, as I suspected from my prior readings of Kadanoff's writings, Kadanoff is not an IDiot. [Kadanoff 2006 quote being unhappy about religious objectors to teaching science snipped] One has to wonder what Kadanoff now thinks of Dembski; especially given Dembski's cowardly role in that trial and in the construction of that "fart video" Dembski put up on the Internet when Judge Jones ruled against the creationist school board.
I am unable to reconcile Kadanoff's 2006 statement with his 2014 one, unless the latter was intended ironically.

... or he was accommodating a former student while being polite but noncommittal. I am not sure that figuring out Kadanoff's motivation or exact attitude is important here, as long as he is not showing up in scientific journals, giving news conferences, or writing blog posts announcing some new scientific argument for Intelligent Design. If readers here know anything about how speakers are chosen for small seminars like this, they will know that Dembski's appearance at the Computations in Science seminar was probably not vetted by the University of Chicago administration, or by its biology departments, or by its mathematical sciences departments, or maybe even by the other faculty members who are involved in the Computations in Science seminar. I know, I run a weekly seminar too.

eric · 6 October 2014

[Quote from UD] Joe Felsenstein chides Dembski for a lack of evidence yet Joe has never presented any evidence for natural selection actually doing something. The point is no one can refute Dembski without providing that evidence. PZ cannot offer up any evidence tat refutes Dembski, that's a certainty

The relevant 'lack of evidence' in this case is the lack of evidence that an intelligent designer is needed to make relatively smooth fitness surfaces. And there isn't just a lack in this case, I think there is good empirical evidence supporting the claim that no designer would be needed. Just to reiterate my previous example, for 64 possible 3-codon sequences to give a white noise type of fitness landscape, a change between any of those 64 sequences must create some significant difference to the phenotype. That is the only way to get a white noise surface. When changes aren't significant, you get a smoother surface. Even without knowing any biology, from a purely statistical perspective one can look at that situation and easily see how the white noise landscape is less likely than a smoother one. The set of codon-to-phenotype translations that would yield a white noise landscape is much smaller than the set of translations that would yield a smooth landscape. Smooth landscapes are in less need of a designer explanation than white noise ones. Dembki's key assumption (i.e., that smoothness needs and explanation because the white noise landscape is some sort of default or null hypothesis) is thus backwards.

Mike Elzinga · 6 October 2014

TomS said: As Harold knows, I am pretty much in agreement with him, and have nothing to cavil with in his latest But I really don't know how much today's fundamentalists are aware of their history.

Indeed, they have their own pseudo-historians - e.g., David Barton - rewriting their history for them. This has become a major industry on the Far Right. They now have Potemkin villages and cargo-cult imitations for just about any academic area you can name; including the "psychology" of gays and female physiology in cases of "legitimate" rape.

Frank J · 6 October 2014

It is thus a Theistic Evolution argument rather than one that argues for Design Intervention.
— Joe Felsenstein

That's the punch line. Everything else is just icing on the cake. In fact the only comment I ever I recall from Dembski as to where and when the design may have been inserted - aside from his usual "It's not ID's task to connect dots" evasion - was that it could have been at the Big Bang. With that he virtually admitted that everything followed "naturalistically" from there. IOW not just "theistic evolution" but specifically "deistic evolution." Ironically, that's too "hands-off" for many "Darwinists," including one of the top critics of ID/creationism of the last 30+ years, Ken Miller. Behe takes a little stronger position than Dembski about the where & when of design insertion, and places it much later, in a designed ancestral cell, but it's essentially as "hands off" as Dembski's. More importantly, no major IDer, including those that seem to doubt common descent, has ever publically challenged that, which suggests private agreement, however reluctant. ID arguments are bogus of course, but the DI will never admit that. So once in a while it's nice to remind them that, even if they were right about finding "design," barring any clear testable alternate "what happened, where, when and how," those results are indistinguishable from that of evolution as we know it, and not remotely like the literal Genesis accounts, young-Earth or old-Earth varieties. The subset of their fan base that is not committed Biblical literalist (most of those prefer AiG and ICR anyway), but hoping for some "evidences," will be interested to hear that.

TomS · 6 October 2014

Frank J said:
It is thus a Theistic Evolution argument rather than one that argues for Design Intervention.
— Joe Felsenstein
That's the punch line. Everything else is just icing on the cake. In fact the only comment I ever I recall from Dembski as to where and when the design may have been inserted - aside from his usual "It's not ID's task to connect dots" evasion - was that it could have been at the Big Bang. With that he virtually admitted that everything followed "naturalistically" from there. IOW not just "theistic evolution" but specifically "deistic evolution." Ironically, that's too "hands-off" for many "Darwinists," including one of the top critics of ID/creationism of the last 30+ years, Ken Miller. Behe takes a little stronger position than Dembski about the where & when of design insertion, and places it much later, in a designed ancestral cell, but it's essentially as "hands off" as Dembski's. More importantly, no major IDer, including those that seem to doubt common descent, has ever publically challenged that, which suggests private agreement, however reluctant. ID arguments are bogus of course, but the DI will never admit that. So once in a while it's nice to remind them that, even if they were right about finding "design," barring any clear testable alternate "what happened, where, when and how," those results are indistinguishable from that of evolution as we know it, and not remotely like the literal Genesis accounts, young-Earth or old-Earth varieties. The subset of their fan base that is not committed Biblical literalist (most of those prefer AiG and ICR anyway), but hoping for some "evidences," will be interested to hear that.

If the end of "design" occurred at the Big Bang, the Origins of Life, or the Cambrian Explosion, that is just a minor variation of Deism. Even, I think, if the last time any Intelligent Designer did anything was at the appearance of Homo sapiens it is a "Clockmaker God" of Voltaire and company.

TomS · 6 October 2014

eric said:
[Quote from UD] Joe Felsenstein chides Dembski for a lack of evidence yet Joe has never presented any evidence for natural selection actually doing something. The point is no one can refute Dembski without providing that evidence. PZ cannot offer up any evidence tat refutes Dembski, that's a certainty
The relevant 'lack of evidence' in this case is the lack of evidence that an intelligent designer is needed to make relatively smooth fitness surfaces. And there isn't just a lack in this case, I think there is good empirical evidence supporting the claim that no designer would be needed. Just to reiterate my previous example, for 64 possible 3-codon sequences to give a white noise type of fitness landscape, a change between any of those 64 sequences must create some significant difference to the phenotype. That is the only way to get a white noise surface. When changes aren't significant, you get a smoother surface.

The relevant 'lack' is a lack of description: what design does so that things are the way they are; what properties of Intelligent Designers have that leads them to the contrivance of design; what gaps in the pre-design world that designed filled. While you note the lack of evidence for the necessity of design, there is also lack of any interest in (let alone reasons to believe) the sufficiency of design. Suppose even that one accepts that God did it, that is not an account for things turning out as they have. That is not an explanation for anything. Even if there had been shown that there was a fatal flaw in evolutionary biology, there is no attempt to show how creation/design repairs that fault.

Doc Bill · 7 October 2014

Over at the Disco Tute site, Evo Whine and Snooze, Dr. Dr. Dumbski writes: "Joe Felsenstein is bad man of straw. Bad straw man, bad! Smoking Joe miss point! Where information come from, huh? All specified and complex, it is."

No, seriously, whiny baby Dembski could post a response right here and mix it up with his antagonists, but, no, he prefers the safety of the Disco Tute crib where someone will change his nappy every hour.

The bottom line from Dembski: 1. Oh, no, you dint! 2. Watch the video. 3. Read my THREE papers. 4. Information! Freedom! Jesus!

eric · 7 October 2014

Doc Bill said: Smoking Joe miss point! Where information come from, huh? All specified and complex, it is."

The point I've been harping on (sorry) is: when we take the general concept of "fitness landscape" and apply it to the way actual, real DNA works, we should recognize that Dembski's white noise landscape is much more specified and complex than alternate landscapes with smooth regions. Its a bit like doing a 1000-run flip of a coin, where we can consider H to be a different fitness than T. In the absence of a designed result, you should see long strings of heads and tails - "smooth" regions. If you don't, if the string is almost entirely a HTHTHT... sequence, then that bespeaks design. A spiky fitness landscape must be designed; in the absence of design, smooth regions are to be expected.

John Harshman · 7 October 2014

I don't think Bill has really thought this through. If it's the fitness surface that's designed rather than the organism itself, that takes a lot more work than just designing species and should be a lot more evident. He's making his case worse, not better. Perhaps he isn't thinking about what the "height" dimension of that surface entails, i.e. the sum of all environmental factors.

Paul C. Anagnostopoulos · 7 October 2014

Dembski has responded:

http://www.evolutionnews.org/2014/10/responding_to_m090271.html

He insists that his arguments don't assume a white noise surface. He also says that a third paper is being ignored:

William A. Dembski, Winston Ewert, and Robert J. Marks II, "A General Theory of Information Cost Incurred by Successful Search," in Marks et al., eds., Biological Information: New Perspectives (Singapore: World Scientific, 2013)

Dieb discusses that paper over at Dieblog.

~~ Paul

Joe Felsenstein · 7 October 2014

I wake up this morning to news of Dembski's response at ENV. I will be responding soon here (most likely with a new post). Meanwhile there's my day job ...

TomS · 7 October 2014

He does assume a law of "conservation" of information.

Has there been any empirical test of that law?

Actually, the proponents of such a law have given us examples of its violation. For example, the sculptures on Mount Rushmore.

Other examples are drawn from the ordinary processes of the world of life.

(These are examples brought up by those who are proponents of such a law, and thus we can be assured that they understand how to calculate the amount of information correctly.)

Therefore, there is no law of conservation of information.

Doc Bill · 7 October 2014

TomS said: He does assume a law of "conservation" of information. Has there been any empirical test of that law? Actually, the proponents of such a law have given us examples of its violation. For example, the sculptures on Mount Rushmore. "Conservation of information" which Dembski totally made up. Just like "complex specified information," "functional complex specified information" and all the other terms the grifter scammers at the Disco Tute make up. What Dembski really means is "Brjekruydm of Wdopkejdke" and it means the same thing. Other examples are drawn from the ordinary processes of the world of life. (These are examples brought up by those who are proponents of such a law, and thus we can be assured that they understand how to calculate the amount of information correctly.) Therefore, there is no law of conservation of information.

Doc Bill · 7 October 2014

Well, as if to prove Dimsky's point, my reply got imbedded in TomS' comment, conservatively speaking. I guess the concept isn't idiot proof!

eric · 7 October 2014

TomS said: He does assume a law of "conservation" of information. Has there been any empirical test of that law?

I have not read Dembski's paper so I may be doing him a disservice, but I suspect he doesn't even give a quantifiable or calculable definition of information, so testing whether it is conserved is several steps away. Which is consistent with what we've seen with Behe and others: their conclusions are irreproducible because they provide no algorithm that independent scientists can use to check their work. And, as Tom points out, it begs the question of what he even means by "conserved" if he and other proponents are already pointing out violations of its conservation. He clearly doesn't mean "conserved" in the way other hard scientists mean it, because for most of us, if humans can increase/decrease quantity X, X isn't a conserved quantity.

eric · 7 October 2014

(...if humans can increase/decrease the total quantity of X in a closed system, I should say...)

Mike Elzinga · 7 October 2014

Dembski keeps dragging the discussion onto his own territory with his misconceptions and misrepresentations of the science behind evolution.

After all these years, Dembski still doesn't understand the underlying point of Dawkins's Weasel program. Dembski, like all ID/creationists, has no clue about the role of physics and chemistry. Neither Dembski nor any other ID/creationist has ever understood physical laws or included physics and chemistry in their calculations. ID/creationism is all about inert things coming together in highly improbable (read impossible) arrangements.

The search-for-a-search issue is totally irrelevant once one understands that physical laws are behind evolution. That is a hurdle ID/creationists will never overcome. Their entire program is based on the crap they inherited from Henry Morris; and they don't even know how it is hobbling them.

Watching ID/creationists going in circles is a bit like watching chickens pacing back and forth in front of a short span of fence and being completely stymied about how to get on the other side.

https://www.google.com/accounts/o8/id?id=AItOawnKupVGX70N9ZsvLu8iScIzWpyVj8bds_Q · 7 October 2014

Massimo Pigliucci has a post up about Stoicism where he relates:

A foundational Stoic precept is to make the distinction between what we can affect (our own attitudes and actions) and what we cannot (external events). Happiness then is available to the person who focuses on what s/he can control. This is achieved by way of the general practice to âfollow nature,â by which the Stoics didnât mean anything like tree hugging or the Paleo diet, but rather both developing the natural human propensity for reason, and accepting that whatever happens is in accord with the way the world works, and it is therefore irrational to oppose it or to become upset by it.

This sums up why ID/creationists are so unhappy - they are trying to change something that can't be changed. How we got here is out of our control.

DiEb · 7 October 2014

W. Dembski says:

Actually, in my talk, I work off of three papers, the last of which Felsenstein fails to cite and which is the most general, avoiding the assumption of uniform probability to which Felsenstein objects.

That seems to be a tad disingenuous: While he hails his paper as one as the "key-stones" of his theories, he doesn't use the model of search which he introduces in it. And for good reason, I think: e.g., in his model most exhaustive searches won't find the target...

DiEb · 7 October 2014

Joe Felsenstein said: [...] The issues I have raised about physics predisposing to smoother-than-random fitness surfaces are separate from these issues of whether (or how) to think about evolution as a search. Dembski had some very unclear statement about evolution being teleological in that it searched for teleology. Me, I'm just searching for the meaning of that statement, and haven't found it yet.

You are not alone: I don't think that Dembski's model - as presented in his talk - is able to describe the evolutionary process (or main parts of it) in a meaningful way.

TomS · 7 October 2014

eric said:
TomS said: He does assume a law of "conservation" of information. Has there been any empirical test of that law?
I have not read Dembski's paper so I may be doing him a disservice, but I suspect he doesn't even give a quantifiable or calculable definition of information, so testing whether it is conserved is several steps away. Which is consistent with what we've seen with Behe and others: their conclusions are irreproducible because they provide no algorithm that independent scientists can use to check their work. And, as Tom points out, it begs the question of what he even means by "conserved" if he and other proponents are already pointing out violations of its conservation. He clearly doesn't mean "conserved" in the way other hard scientists mean it, because for most of us, if humans can increase/decrease quantity X, X isn't a conserved quantity.

I wonder (I really do wonder, and think that it is an interesting question) when Wohler synthesized urea, and, so we are told, provided proof that there was no essential difference between organic and inorganic chemstry - were there people who did not accept this a conclusive demonstration because Wohler intervened in the sythesis. Wohler being, so to speak, an organic agent (or catalyst) in the reaction. I'm pretty sure that when the intelligent designers of particle accelerators produced violations of charge-parity conservation that that was accepted as a demonstration that there was no law of conservation of charge-parity. As far as we know, intelligent designers are as subject to the laws of nature as everything is. The default assumption should be that if there is a violation of a proposed law, that that means that there is no such law.

TomS · 7 October 2014

https://www.google.com/accounts/o8/id?id=AItOawnKupVGX70N9ZsvLu8iScIzWpyVj8bds_Q said: Massimo Pigliucci has a post up about Stoicism where he relates:
A foundational Stoic precept is to make the distinction between what we can affect (our own attitudes and actions) and what we cannot (external events). Happiness then is available to the person who focuses on what s/he can control. This is achieved by way of the general practice to âfollow nature,â by which the Stoics didnât mean anything like tree hugging or the Paleo diet, but rather both developing the natural human propensity for reason, and accepting that whatever happens is in accord with the way the world works, and it is therefore irrational to oppose it or to become upset by it.
This sums up why ID/creationists are so unhappy - they are trying to change something that can't be changed. How we got here is out of our control.

And it sums up why "evolutionists" are so unhappy - they are trying to use reason to change minds that can't be changed by reason.

Rolf · 9 October 2014

TomS said:

And it sums up why âevolutionistsâ are so unhappy - they are trying to use reason to change minds that canât be changed by reason.

That, in a nustshell, is all that it is all about. A prerequisite is the ability to listen to your opponent. I may be wrong but to me it looks like Dembski is arguing from his ivory tower like a muezzin from his minaret, not really engaging in discourse to sort out any disagreement. Lost causes are hard to defend. Makes me think of Ruhollah Khomeini or Muammar al-Gaddafi.

harold · 9 October 2014

As far as we know, intelligent designers are as subject to the laws of nature as everything is.

One of the fundamental logic errors of ID is false analogy to human design. The "forensic scientists and archaeologists detect design" crap line was the go-to line for trolls for years. Of course, it's the opposite. Whatever their private beliefs, competent forensic scientists and archaeologists never ascribe their findings to miraculous supernatural causes, and could not function at all if they did not favor natural explanations. Also, of course, anything can be analogized to human design. Funny looking rock? Looks like an abstract sculpture. The "is there anything that isn't 'designed' and how can we tell the difference?" question is always valuable.

Joe Felsenstein · 9 October 2014

Rolf said: TomS said:
And it sums up why âevolutionistsâ are so unhappy - they are trying to use reason to change minds that canât be changed by reason.
That, in a nustshell, is all that it is all about. A prerequisite is the ability to listen to your opponent. I may be wrong but to me it looks like Dembski is arguing from his ivory tower like a muezzin from his minaret, not really engaging in discourse to sort out any disagreement. Lost causes are hard to defend. Makes me think of Ruhollah Khomeini or Muammar al-Gaddafi.

Wait a minute -- you're accusing Dembski of murdering how many people?

Frank J · 9 October 2014

If the end of âdesignâ occurred at the Big Bang, the Origins of Life, or the Cambrian Explosion, that is just a minor variation of Deism. Even, I think, if the last time any Intelligent Designer did anything was at the appearance of Homo sapiens it is a âClockmaker Godâ of Voltaire and company.
— TomS

Which means that he/she/it/they didn't do it then, but "programmed" it, billions of years earlier. And, as Behe admitted at Dover, the designer(s) could have been no longer in existence by then. To me the other half of the game was given away by PT's own FL ~6 years ago. Probably inspired by your frequent mention of reproduction, I asked his opinion if human reproduction was a design actuation event. I expected evasion, but was surprised that he said yes without hesitation. In retrospect I didn't think of specifying whether the design was inserted in real time or programmed in advance, but the point is moot, since his was an effective admission that what we observe (chance and regularity) and "design" are not mutually exclusive. Thus undermining not only the DI's pretense that they are mutually exclusive, but also their unwritten policy to lead the reader to infer that design actuation occurred "long ago, but don't ask where when or how."

Frank J · 9 October 2014

I wonder (I really do wonder, and think that it is an interesting question) when Wohler synthesized urea, and, so we are told, provided proof that there was no essential difference between organic and inorganic chemstry - were there people who did not accept this a conclusive demonstration because Wohler intervened in the sythesis. Wohler being, so to speak, an organic agent (or catalyst) in the reaction.
— TomS

If by "people" you mean other chemists, I don't know if any used it as a "design" argument (good question, though!), but I recall reading that some insisted that it could not be the same substance as the one produced biologically (i.e. a "vitalism" argument). Though not for long, as the list of synthesized organic substances skyrocketed. OTOH, many (most?) nonscientists even now think there's a fundamental difference between (what they call) "organic" and "inorganic" substances. There's a big overlap between the "natural and organic" crowd and evolution-deniers. I often hear the incredulity of abiogenesis (and confusion with evolution) stated as: "I can't believe that a bunch of inorganic chemicals can suddenly become a living thing." To which I would chime in, "Actually it was organic substances (*) that assembled into a living system." (*) Another of my pet peeves is the use of the word "chemical" as a noun. As with "natural" and "organic" it feeds public misconceptions about science.

Rolf · 9 October 2014

Anyone familar with the word "Google" can find answers and verify that their own concepts are relevant. Wikipedia:

An organic compound is any member of a large class of gaseous, liquid, or solid chemical compounds whose molecules contain carbon.

fnxtr · 9 October 2014

Okay, so.. the world is billiard balls knocking each other around on a bed of nails and life is the chance that they all balance on the points of the nails. Right, Bill?

TomS · 9 October 2014

The inorganic chemicals dihydrogen monoxide and that compound containing poison chlorine and sodium are often found in one's diet, and become part of living cells.

harold · 9 October 2014

Thereâs a big overlap between the ânatural and organicâ crowd and evolution-deniers.

I'm not sure that's the case. Evolution denial has two main demographics. Active evolution denial is a right wing political thing, and that can't be stressed enough, because it can't be addressed if it isn't understood. Passive evolution denial, that is, personal willingness to answer a poll question about evolution wrong and maybe belong to a religious denomination that is at least passively "Biblically literalist", without agitating for illegal public school curricula, is more diverse, and is more associated with low income and low education (while including all income levels, just more slanted toward low income). Everywhere outside of the deep south, including the rest of the south, low income demographics are more Democratic in slant than ethnically similar higher income demographics, so this group would be expected to be less right wing overall. (The quintessential right wing demographic is absolutely NOT low income white, which is not a very right wing demographic overall, it is low education/high income white.) However, although I wish this group of people would learn more about science and understand that they can get the psychological benefits of religion without denying science, unlike the group described in the first paragraph, these people aren't hurting anyone. When I say "creationist", not otherwise specified, I mean the politically active type. Organic and natural types, a group I could be accused of being a borderline member of, do indeed sometimes make absurd, unsupported claims. They are usually well off and usually not hard right wing, so they don't fit either evolution denial demographic. And when they are being unscientific, they usually make unsupported claims, or express unwarranted fears, about details, which is, although silly very different from outright denying a major scientific theory. There is some level of science confusion among almost everyone in the population. Even scientists are sometimes confused about the work of other scientists. Organic farming is not, in itself, a science denying or irrational activity. There are advantages to the use of industrial agricultural chemicals, and there are advantages to limiting their use. It depends on the context. I have a cousin who is an organic farmer. It wasn't a very hard decision for him. His family has been raising grass fed beef for generations and he inherited the beef farm; the grass was already organic. He told me he used some "homeopathic" remedy for some kind of allergy symptoms. I thought it must be placebo, but I turned out to be wrong. The product is legally marketed as "homeopathic" but actually includes a reasonable concentration of a well known active medication. It wasn't that homeopathy works but that the product isn't really homeopathic. As far as I know they could market processed cheese flavored doritos as "homeopathic" if they wanted, the word "homeopathic" isn't regulated.

harold · 9 October 2014

Rolf said: Anyone familar with the word "Google" can find answers and verify that their own concepts are relevant. Wikipedia:
An organic compound is any member of a large class of gaseous, liquid, or solid chemical compounds whose molecules contain carbon.

"Organic" has several meanings. When I took my "organic chemistry" classes, they did indeed deal with the fascinating chemistry of carbon based compounds. There was a good deal of overlap between those classes and "biochemistry", of course. Organic can also imply biological. And it also has a legally regulated meaning in agriculture. There's nothing inherently wrong or stupid about sometimes raising some types of crops without use of industrial agricultural chemicals. Current agricultural practices aren't very sustainable. We can't just make everything "organic", but using organic methods as part of a mix of less polluting, less reliant on non-newable resources, more sustainable agriculture is reasonable.

TomS · 9 October 2014

I remember when I first saw a section of my local supermarket announcing "organic foods". It stirred me out of my mindless shopping state as I tried to figure what the inorganic foods that the rest of the store was selling were.

harold · 9 October 2014

TomS said: I remember when I first saw a section of my local supermarket announcing "organic foods". It stirred me out of my mindless shopping state as I tried to figure what the inorganic foods that the rest of the store was selling were.

Food items can technically be "organic" in an agricultural sense and "inorganic" in a chemical sense, for that matter. Someone could probably sell an inorganic edible salt of calcium (chemically inorganic, that is) as an organic calcium supplement, if it was prepared from USDA certified organic ingredients. However, whatever anyone's opinion of organic foods, I do want to nip in the bud any false equivalence between things like exaggerated fear of gluten and whatnot, versus creationist activities like not only directly denying major established scientific theories, but doing so in a very dishonest way, making false accusations against scientists, and agitating to have sectarian science denial taught as "science" in public schools. I can think of all kinds of organic versions of junk based on processed flour, sugar, and so on. That's somewhat silly. You might as well buy the regular fudgee-o's to all extents and purposes. Organic white sugar still promotes obesity and tooth decay when consumed in excess, and is ideally avoided. But eating organic fudgee-o's isn't quite the equivalent of scheming to destroy science education because your ideology is inconvenienced by science.

https://me.yahoo.com/a/JxVN0eQFqtmgoY7wC1cZM44ET_iAanxHQmLgYgX_Zhn8#57cad · 9 October 2014

TomS said: I remember when I first saw a section of my local supermarket announcing "organic foods". It stirred me out of my mindless shopping state as I tried to figure what the inorganic foods that the rest of the store was selling were.

Salt. I guess they were pointing out that this food wasn't salt. Actually, I assume that the "organic" label for food goes back to older meanings of "organic," like that it supposedly has a more "organic" relationship to earth, life, and nature, with natural cycles of nutrients instead of with all of the mined and manufactured inputs and "unnatural" pesticides. Since it's often highly-intensive ag, I'm not sure that even that meaning truly fits, but I assume that it's comforting to many who buy the food, and the pesticides certainly can be an issue. One thing that's true is that it's the "organic" in chemistry that has seriously strayed from the original meaning, which to many meant a kind of vitalism. It was a big deal when something "organic," urea, was actually synthesized in a lab using "inorganic" chemicals. Now "organic" in chemistry means carbon chemistry, with a relatively few exceptions (CO2, carbonates, elemental carbon, etc.), but once it was a mysterious realm that was thought not to reduce down to the normal rules of chemistry. Glen Davidson

Rolf · 9 October 2014

Joe Felsenstein said:

Wait a minute -- you're accusing Dembski of murdering how many people?

A rhetorical reply, but anyway, my comparison did of course not extend beyond the nature of the misanthropic mind of those characters. How they used it is entirely another matter.

George Frederick Thomson Broadhead · 10 October 2014

This comment has been moved to The Bathroom Wall. I warned Broadhead -- his comments are off-topic entirely. JF

George Frederick Thomson Broadhead · 10 October 2014

This comment has been moved to The Bathroom Wall I warned Broadhead -- his comments are entirely off-topic. JF .

Joe Felsenstein · 10 October 2014

Rolf said: Joe Felsenstein said:
Wait a minute -- you're accusing Dembski of murdering how many people?
A rhetorical reply, but anyway, my comparison did of course not extend beyond the nature of the misanthropic mind of those characters. How they used it is entirely another matter.

Good that you have qualified that. I realize people have had a hard day at work and like to relax by calling names. But watch it with accusing people of crimes. That one came very close to lawsuit territory, and PT would be at risk in such a case. If I see other comments that accuse people of crimes I will delete them as sending them to the Wall still leaves us potentially liable.

Rolf · 10 October 2014

Joe Felsenstein said:
Rolf said: Joe Felsenstein said:
Wait a minute -- you're accusing Dembski of murdering how many people?
A rhetorical reply, but anyway, my comparison did of course not extend beyond the nature of the misanthropic mind of those characters. How they used it is entirely another matter.
Good that you have qualified that. I realize people have had a hard day at work and like to relax by calling names. But watch it with accusing people of crimes. That one came very close to lawsuit territory, and PT would be at risk in such a case. If I see other comments that accuse people of crimes I will delete them as sending them to the Wall still leaves us potentially liable.

I must clarify. The previous was written hastily at the end of the day here and later I realized that what I'd actually had in mind when I wrote my first comment was just the impression I always got when watching them on TV: Stern, unmoving, impervious to arguments. Nothing else, not even misanthropy was implied.

https://me.yahoo.com/a/JxVN0eQFqtmgoY7wC1cZM44ET_iAanxHQmLgYgX_Zhn8#57cad · 10 October 2014

Joe Felsenstein said:
Rolf said: Joe Felsenstein said:
Wait a minute -- you're accusing Dembski of murdering how many people?
A rhetorical reply, but anyway, my comparison did of course not extend beyond the nature of the misanthropic mind of those characters. How they used it is entirely another matter.
Good that you have qualified that. I realize people have had a hard day at work and like to relax by calling names. But watch it with accusing people of crimes. That one came very close to lawsuit territory, and PT would be at risk in such a case. If I see other comments that accuse people of crimes I will delete them as sending them to the Wall still leaves us potentially liable.

Oh come on, that wasn't even close to accusing anyone of a crime, nor close to actionable as libel. Hyperbolic, certainly, arguably beyond what should be said, but also tied to the "lost causes" and denial, not to crimes against humanity or any such thing. "Makes me think of" is ambiguous, and potentially covers a whole lot, yet it definitely doesn't amount to much of an actual accusation, especially not one of crime. Your rules and all. But it was a bit of hyperbole, at worst. Glen Davidson

Frank J · 10 October 2014

Organic and natural types, a group I could be accused of being a borderline member of, do indeed sometimes make absurd, unsupported claims. They are usually well off and usually not hard right wing, so they donât fit either evolution denial demographic.
— Harold

I personally know at least one in the "active" group. But I think you underestimate the size and variety of the "passive" group. Many of them say such things as "I hear the jury's still out about evolution," and are willing to uncritically believe, and repeat, anything negative said against any kind of science. Politically and religiously they're all over the map. As you know, the active deniers (anti-evolution activists) and the subset of "passive" that are committed Biblical literalists are almost all far-right authoritarians, which means not necessarily the most conservative on economic issues.

harold · 10 October 2014

Joe Felsenstein said:
Rolf said: Joe Felsenstein said:
Wait a minute -- you're accusing Dembski of murdering how many people?
A rhetorical reply, but anyway, my comparison did of course not extend beyond the nature of the misanthropic mind of those characters. How they used it is entirely another matter.
Good that you have qualified that. I realize people have had a hard day at work and like to relax by calling names. But watch it with accusing people of crimes. That one came very close to lawsuit territory, and PT would be at risk in such a case. If I see other comments that accuse people of crimes I will delete them as sending them to the Wall still leaves us potentially liable.

While it would be exceptionally difficult for anyone to successfully sue a web site over a nonspecific comparison to a historical figure in the comments section, I agree with this policy on a different grounds. False accusations and exaggerated metaphors are the characteristic techniques of those who lack a viable argument. But an unfair accusation immediately creates sympathy for the target. That sympathy may actually shield the target from deserved criticism for their real obnoxious characteristics. For example, above, I noted, in an on topic context, some disadvantages of eating too much refined sugar. I could also note that sugary foods are heavily marketed, for obvious commercial reasons. If I were to go on a rant comparing "Big Sugar" to Ivan the Terrible, on the other hand, and making false accusations of crimes against sugar cane growers, my valid points points about refined sugar would be diluted by the obvious unreasonable nature of my subsequent comments. Now, having said that, I do agree that to me William Dembski exhibits, at least in his works and public appearances, an unyielding, often disrespectful, often insulting, often defensive character, and that in these ways, he does resemble authoritarian personality types (as I perceive them) from other realms of endeavor, including politics. However, comparisons to bloodthirsty tyrants are unfair, since he hasn't committed any known violent crimes. It's obvious that the comparison meant only to refer to their non-violent personality traits, but it's too ambiguous, and someone could misread it as an accusation of similar violent crimes.

harold · 10 October 2014

Frank J said:
Organic and natural types, a group I could be accused of being a borderline member of, do indeed sometimes make absurd, unsupported claims. They are usually well off and usually not hard right wing, so they donât fit either evolution denial demographic.
— Harold
I personally know at least one in the "active" group. But I think you underestimate the size and variety of the "passive" group. Many of them say such things as "I hear the jury's still out about evolution," and are willing to uncritically believe, and repeat, anything negative said against any kind of science. Politically and religiously they're all over the map. As you know, the active deniers (anti-evolution activists) and the subset of "passive" that are committed Biblical literalists are almost all far-right authoritarians, which means not necessarily the most conservative on economic issues.

Agreed that a broad definition of evolution denial captures a diverse population - although that population is enriched in the two demographics I noted. I'm not sure that "natural and organic" types overlap with evolution denial. I might expect them to be, while sometimes having their own issues with some aspects of science, far more likely to accept evolution, at least passively, than the general population. The typical consumer of organic products is probably more affluent and more educated than the average person, and less likely, relative to income, to be right wing in politics. And probably much less likely to be authoritarian, what right wing tendencies there are in this group slant toward callous libertarianism. If you look at the controversial science related ideas associated with "hippies" circa 1970, you see that some of them were clearly wrong and silly, but others were, whether or not for a good reason, prescient. That isn't the case with authoritarian creationists. "Creation science" from 1970 is just 100% grotesquely wrong.

shjcpr · 12 October 2014

I think Dembski has made it clear who really understands the mathematics behind evolution.

The argument from labor and authority (only a biologist is qualified to discuss biology) no longer packs the rhetorical punch is may once have had. Its pretty lame to try rebutting Dembski's argument by claiming it doesn't represent 'real' evolution.

One scotsman is too many. Nothing against Scots, mind you.

Besides, one ought to be rather skeptical of 'mindless' arguments, which ironically can only be detected by 'mindful' inquiry.

Down the rabbit hole you go, mind-lessness.

Mike Elzinga · 13 October 2014

shjcpr said: I think Dembski has made it clear who really understands the mathematics behind evolution.

Indeed, the contrast couldn't be starker; Dembski loses and real scientists win at every turn. There are no "rhetorical punches" in real science. Science isn't done by staged debates; people doing science really do have to know something. No one in the science community believes that Scrabble letters, dice, coins, and inert objects are stand-ins for the behaviors of atoms and molecules. No calculations in chemistry, physics, or biology look anything like Dembski's calculations. Calculations in science are actually about the interactions among real things. Dembski doesn't know about interactions or condensed matter; his Chicago talk demonstrated that as clearly as anything could. Dembski doesn't understand science at any level; and he certainly doesn't do any science. In fact, Dembski isn't able to get beyond logarithms to base two of the probabilities of the arrangements of inert objects. Even high school students can do better. But perhaps as an expert yourself, you can explain what Dembski's methods of calculating have to do with real atoms and molecules in the real world of physics, chemistry and biology.

shjcpr · 13 October 2014

Well Frank and Harold, it would be interesting enough to do a survey of design deniers like yourself.

...pedantic tendencies, secular philosophical bent, extreme extrapolations from scant empirical evidence, conflate technological advances with scientific advances....

"Design denial - the flat-earth syndrome of the 21st century"

Mike Elzinga · 13 October 2014

It appears that we have here a version of Joe G or Mung the Magnificent feces flinger popping in from over at UD to throw taunts here at Panda's Thumb. Maybe a Steve P?

Anybody see a familiar MO here?

shjcpr · 13 October 2014

Well Mike, if it was all physics and chemistry, even you would have it licked by now.

Insanity is repeating the same-o over and over expecting a different result. So why do you keep insisting its all physics and chemistry but we keep getting the same result. Nothing. So how long are we supposed to wait Mike? Another 100 years, another 1000???

Even Leo Kadanoff understands that much.

At least Dembski is pushing the envelope in the right direction, trying to break the fruitless design denying cycle of failure.

Yet, Mike would not have it any other way.

Physic...chemistry....all the way down.

Mike Elzinga · 13 October 2014

shjcpr said: Well Mike, if it was all physics and chemistry, even you would have it licked by now.

In fact I and many others have explained it; but you never even got past that little thermodynamics concept test. Crawled away you did. Are you going to explain Dembski's calculations for us? We know you can't.

Dave Luckett · 13 October 2014

Writes like Stevie P the Taiwanese rug merchant. Barely coherent fact-free invective that has the effect of expressing generalised scorn for reality.

Mike Elzinga · 13 October 2014

Dave Luckett said: Writes like Stevie P the Taiwanese rug merchant. Barely coherent fact-free invective that has the effect of expressing generalised scorn for reality.

That's what I suspect also. Shouldn't be too long before he gets shipped off to the Bathroom Wall where he can peddle his rugs in pieces. I won't be responding to him any further; I think this short exchange has been enough to identify him.

shjcpr · 13 October 2014

Reality gets under Luckett's skin.

He prefers the comfort of mindlessness.

Its easy to pawn off the difficulties onto emergence.

While design stares you right in the face.

Man designs. Where did he get that capability? From nature. How? Well, because nature designs. In fact nature knew long ago all of the design principles Man recently 'invented'. How do we know?

All you have to do is follow the money.

shjcpr · 13 October 2014

thats what you expected???? It too easy to find out that it is in fact Steve P.

And I guess Luckett's darwinian bent lets fact evolve into fiction. But he didnt do that on purpose. NO. Couldn't accuse him of a lie. He's just parrotting Stanton is all. No harm done, right.

shjcpr · 13 October 2014

Of course Mike has no good answer. Just the stock responses. No change here.

Joe Felsenstein · 13 October 2014

[Voice echoes from the sky while lightning bolts flash]

Although shjcpr may well be one of our regular trolls, a quick look at IP numbers finds no match (which doesn't mean it isn't one of our regular trolls).

I don't have the time or energy to move all this exchange to the Bathroom Wall, But further content-free gibes by shjpcr will not remain here and will go to the BW. As will replies to them.

If shjcpr wants to explain, in scientific detail, how Dembski's argument is supported and his critics in this thread get it wrong, that will be allowed. But just arguing that criiticism here is "stock responses", "mindlessness", "design denial" is Bathroom Wall material.

There has to be an actual argument in the comment or shjpcr's declarations of victory will go there. Without scientific detail they are the usual stupid trollery.

RPST · 13 October 2014

This comment has been moved to The Bathroom Wall. I meant it, folks. One lightning bolt thrown, many more in my quiver. JF

apokryltaros · 13 October 2014

So, Dembski is coughing up another batch of word-hairballs, this time physics-flavored, and still can't be bothered or trusted to explain how his pseudo-explanation simultaneously prevents evolution from occurring and permits people to detect Intelligent Design (r) ?

*yawn*

nmanningsam · 17 October 2014

This comment has been moved to The Bathroom Wall as will happen to all replies to the contentless trolling by shjcpr. JF

eric · 17 October 2014

This comment has been moved to The Bathroom Wall as will happen to all replies to the contentless trolling by shjcpr. JF.

TomS · 17 October 2014

This comment has been moved to The Bathroom Wall as will happen to all replies to the contentless trolling by shjcpr. JF.