Bankrupting ID
Jamie Bridgham, Sean Carroll and Joe Thornton of the University of Oregon decided it was high time someone discovered a Darwinian mechanim producing irreducible complexity, and so they set to work on the problem. From Thornton’s website:
Molecular evolution of hormones and their receptors
How did hormones and their diverse functions in humans and other animals evolve? We study the evolution of vertebrate steroid hormones — such as estrogen, testosterone, and the stress hormone cortisol — and the receptor proteins that mediate these hormones’ effects on the body’s cells. Our goal is to reveal the specific molecular events by which hormones and receptors diversified and evolved their specific partnerships. By combining techniques from statistical phylogenetics, molecular endocrinology, ancestral gene resurrection, and experimental evolution, we are characterizing receptor biodiversity across the animal kingdom, reconstructing the evolution of the family at the genetic level, and testing hypotheses about the functions of ancient genes. Our goal is to illustrate how a complex, tightly integrated molecular system — one which appears to be "irreducibly complex" - evolved by Darwinian processes hundreds of millions of years ago.
Higly interesting work, and their results were published in Science this week– along with a considerable amount of fanfare. The University of Oregon put out a press release:
And the New York Times picked up the story. It was going wonderfully, but there was one slight catch.
Note to potential researchers: if you want to falisify irreducible complexity, it helps to work with an irreducibly complex system.
Behe responded to the article, labeling it as "the lamest attempt yet to answer the challenge irreducible complexity poses for Darwinian evolution". According to him…
The bottom line of the study is this: the authors started with a protein which already had the ability to strongly interact with three kinds of steroid hormones (aldosterone, cortisol, and “DOC” [11-deoxycorticosterone]). After introducing several simple mutations the protein interacted much more weakly with all of those steroids. In other words, a pre-existing ability was decreased.
…
The authors (including Christoph Adami in his commentary) are conveniently defining “irreducible complexity” way, way down. I certainly would not classify their system as IC.
…
In conclusion, the results (and even the imagined-but-problematic scenario) are well within what an ID proponent already would think Darwinian processes could do, so they won’t affect our evaluation of the science. But it’s nice to know that Science magazine is thinking about us!
The fellows of the DI Center for Science and Culture have a more thorough analysis of "what went wrong" in Bridgham et al’s study here.
Perhaps the paper deserves a bit more careful analysis.
In the latest curious development, Joe Thornton seems to have decided he never meant to test irreducible complexity after all. Compare the cached and current versions of his webpage here and here. The quote above is from the original version.
It doesn’t seem that Google’s cache reflects the change any longer. Looks like they’ve updated their search engine indexing. Oh well. Does anyone know how we can get an older cached copy of Thornton’s site?
Comment by Wiglaf — April 10, 2006 @ 1:49 pm
Of course, Behe’s response ignores the fact that the Bridgham et al. paper uses the same example of IC that Behe himself gave on the witness stand in Dover (namely, a protein binding site). You can read the actual story at the Panda’s Thumb.
Comment by Andrew — April 10, 2006 @ 4:27 pm
Andrew–
Thanks for the link to the Panda’s Thumb article. Do you have a citation for where in the Dover Trials Behe declared all protein binding sites irreducibly complex? From what I know of his work a general statement of that sort seems strangely out of character.
Wiglaf–
I took a screenshot; I guess I’ll link to that.
Comment by Freawaru — April 10, 2006 @ 4:38 pm
Dover trial day 11 (Q: = prosecutor, A = Behe):
“Q And there are zero articles in peer-reviewed scientific journals arguing for the irreducible complexity of complex molecular systems?
A There are none that use that phrase, but as I indicated in my direct testimony, that I regard my paper with Professor David Snoke as to be arguing for the irreducible complexity of things such as complex protein binding sites.”
http://www.talkorigins.org/faqs/dover/day11pm2.html#day11pm547
Link to abstract for Behe & Snoke paper:
http://www.proteinscience.org/cgi/content/abstract/ps.04802904v1
“Yet some protein features, such as disulfide bonds or ligand binding sites, require the participation of two or more amino acid residues, which could require several mutations.” [Goes on to conclude population size of 10 ^ 9 required for such mutations]
Comment by Dizzy — April 10, 2006 @ 4:51 pm
I’d be hard pressed to draw the conclusion that Behe regards all “complex protein binding sites” to be irreducibly complex from that excerpt. Is there more background I should be considering?
Comment by Freawaru — April 10, 2006 @ 5:07 pm
Freawaru, how about this April 2002 address by Behe to the American Museum of National History (and hosted on the DI’s website — at least for the time being):
“The problem of irreducibility in protein features is a general one. Whenever a protein interacts with another molecule, as all proteins do, it does so through a binding site, whose shape and chemical properties closely match the other molecule. Binding sites, however, are composed of perhaps a dozen amino acid residues, and binding is generally lost if any of the positions are changed. One can then ask the question, how long would it take for two proteins, that originally did not interact, to evolve the ability to bind each other by random mutation and natural selection, if binding only occurs when all positions have the correct residue in place?”
For Behe to argue that the exact same process is now “piddling” (because it disproves his assertions!) is precisely the same combination of arrogance and stupidity that so poorly served the ID movement on the stand in Kitzmiller.
There are only two options here: either (1) Bridgham et al. have conclusively shown that naturalistic evolutionary pathways can and do give rise to IC systems; or (2) the definition of IC is inherently circular, question-begging, and worthless, because Behe and the ID folks will simply move the goalposts every time real scientists do real work to demonstrate that ID assertions are false.
Comment by Andrew — April 11, 2006 @ 1:31 am
Let’s look at that quote again:
“One can then ask the question, how long would it take for two proteins, that originally did not interact, to evolve the ability to bind each other by random mutation and natural selection, if binding only occurs when all positions have the correct residue in place?”
I was under the impression that the original proteins that they were proposing did in fact already interact with the binding site, thus not meeting the definition.
Also, just to point out (I don’t know if this has any affect on the current paper or not), Behe in IC systems was explicitly talking about _random_ mutations. IC can evolve by nonrandom mutations.
Comment by Jonathan Bartlett — April 11, 2006 @ 4:32 am
Jonathan: Behe was talking about evolutionary pathways in that paragraph; thus, to become a binding site, two proteins that “originally did not interact” would have to come to interact — something he claims is IC and evolutionarily impossible.
All proteins interact with their binding sites; that’s what “bind” means. So your impression is incorrect.
Comment by Andrew — April 11, 2006 @ 3:37 pm
“Behe was talking about evolutionary pathways in that paragraph; thus, to become a binding site, two proteins that “originally did not interact” would have to come to interact”
Right. My point was that in the paper that supposedly debunked Behe I think the binding sites were already there. However, I haven’t made it to the library yet to check the paper. Perhaps I will do it today at lunch.
According to Behe:
So, not only did they attack the weakest form of IC (Behe’s book is about systems much more complex involving numerous proteins), but it seems they weren’t even successful at it.
“something he claims is IC and evolutionarily impossible.”
Actually, as I pointed out in my previous post, the claim is that it’s evolutionarily impossible by _Darwinian_ evolution.
Comment by Jonathan Bartlett — April 11, 2006 @ 3:54 pm
Then on sensitivity in http://www.pandasthumb.org/archives/2006/04/evolution_of_ic_1.html#comment-96005 :
“Behe wrote: “The authors do not test for that [that S106P would be weeded out by natural selection]; they simply assume it wouldn’t be a problem, or that the problem could somehow be easily overcome. Nor do they test their speculation that DOC could somehow act as an intermediate ligand. In other words, in typical Darwinian fashion the authors pass over with their imaginations what in reality would very likely be serious biological difficulties.”
This shows the limitations of critiquing biology when you don’t know too much about it. Now, Thornton at least, being a steroid receptor specialist (and most of his receptor specialist readers) would be aware of the plasma steroid hormone levels and hence know he was talking about effectively neutral mutations. Furthermore, if you look at figure 2B in the BCT paper, you will see that the standard Teleost glucocorticoid receptor has effectively the same cortisol sensitivity as the S106P receptor. In fact, the Teleost receptor a bit less sensitive, and a fair bit less effective, than the S106P receptor, and fish do just fine. So for all intents and purposes, the S106P mutation is neutral.
Behe is very unlikely to know what the plasma levels of these steroids are, but he should have had a look for them if he had any biological background, as the plasma levels of a hormone are critical for understanding how the receptor functions. Also, he should have looked a wee bit more closely at look closely at figure 2B, showing the very low Teleost sensitivity to cortisol (the significance of which, if you are not a receptor biologist like me, is not immediately apparent). Behe assumed BCT pulled neutrality out of a hat, but we know from the biology that the S106P mutation will be neutral (after all, fish function quite nicely with a “worse” receptor).”
So we have an increase of 3000 times in selectivity between the ancestral receptor to the evolved GR receptor, with more than enough sensitivity. But Behe doesn’t seem to understand that.
“Actually, as I pointed out in my previous post, the claim is that it’s evolutionarily impossible by _Darwinian_ evolution.”
The paper that is refered to discusses neo-darwinistic mechanisms acting on the genome. I don’t see the point in making a distinction between the basic RM + NS mechanism and neo-darwinism here. Also, to say that IC, which has never been proved to exist but has many argument against it, evolution being only one of many, and is nothing but a negative argument *against* evolution, is explainable by neo-darwinistic evolution seems… odd.
Comment by Torbjörn Larsson — April 11, 2006 @ 9:29 pm
Seems the first half of a very long comment got deleted.
““Behe was talking about evolutionary pathways in that paragraph; thus, to become a binding site, two proteins that “originally did not interact” would have to come to interact”
Right. My point was that in the paper that supposedly debunked Behe I think the binding sites were already there.”
A binding site were there in the ancestral receptor, but not the final MR site that binds to tetrapod aldosterone. There was no original interaction, since aldosterone wasn’t yet developed in any synthesis pathway. This shows the development of a new hormone-receptor pair.
We can also note in passing that Behe’s statement “Binding sites, however, are composed of perhaps a dozen amino acid residues, and binding is generally lost if any of the positions are changed.” is refuted by this paper. Two residues were changed with changes in specificity but no loss.
“So, not only did they attack the weakest form of IC (Behe’s book is about systems much more complex involving numerous proteins), but it seems they weren’t even successful at it.”
This is adressed by Ian Musgrave in two comments on The Panda’s Thumb.
First on selectivity in http://www.pandasthumb.org/archives/2006/04/evolution_of_ic_1.html#comment-95979 :
“Behe wrote: “Nothing new was produced in the experiment; rather, the pre-existing ability of the protein to bind several molecules was simply weakened. The workers begin their experiments with a protein that can strongly bind several, structurally-very-similar steroids, and they end with a protein that at best binds some of the steroids ten-fold more weakly. (Figure 4C)”
This is called producing an increase in selectivity. Producing selective receptors is rather important from an organisms point of view[1]. This is something that fellow ID travellers like Lee Spetner says can’t happen (yet it does, see my critique of Spetner’s arguments for more information about the evolution of binding sites). Rather than being activated by multiple steroids, the receptor is now more effectively activated by cortisol than aldosterone (just like the modern receptor, lets not forget that these mutations produce a receptor with the sensitivity and selectivity of the modern receptor). The evolved receptor has a thousand fold less affinity for aldosterone than the ancestral one, while only 10 fold less affinity for cortisol. The evolved receptor is 100 times more selective for cortisol than aldosterone (whereas the ancestral one was 30 fold more selective for aldosterone). Behe obscures this important point by citing only the 10 fold change. Furthermore, the evolved receptor is sensitive enough to cortisol that it will be effectively activated by the plasma cortisol levels found in teleosts and tetrapods; neither so sensitive that is permanently jammed on by these levels, but can respond to stress-induced changes, nor so insensitive that it will not be activated (and basically has the same sensitivity as the modern GR). These are important points, although not actually relevant to IC, that Behe either ignores or doesn’t understand in his attempt to minimise the importance of the BCT paper.”
And when continue with the first comment.
Comment by Torbjörn Larsson — April 11, 2006 @ 9:51 pm
But this sounds like a pathway that would not be irreducibly complex. Behe has never claimed that variations or even improvements in systems are counterarguments to irreducible complexity.
Again, the binding site was _already there_.
“This is called producing an increase in selectivity.”
Nowhere that I am aware of does Behe list increase or decrease in selectivity as being irreducibly complex.
“Producing selective receptors is rather important from an organisms point of view”
Irrelevant if Behe has not claimed it to be irreducibly complex. The IC argument is _not_ that nothing can evolve by Darwinian mechanisms, but that _certain_ things cannot evolve by Darwinian mechanisms. Nowhere have I found that Behe regards increases or decreases in selectivity to be irreducibly complex changes.
“This is something that fellow ID travellers like Lee Spetner says can’t happen”
Irrelevant what Spetner says, though I would be interested in seeing the reference.
And, on top of all of this, it only produces selective advantage in the context of a larger system. So there is no reason to think these changes would be of any use at all (and therefore no reason to think they would be conserved) except in the light of a larger system which needs them, in which case we would need to evaluate how it would or would not be selected.
And, again, remember that Behe only claims non-evolvability against _Darwinian_ mechanisms, not other types of mechanisms. As well, this sort of system is orders of magnitude less complex than those he describes in his book. Also, Behe has only claimed evolution of _new_ binding sites as IC, not the repurposing of old ones (the Behe and Snokes paper was about a _new_ binding site).
I don’t see how you can try to use Behe as who you are arguing against when you don’t even use his own definitions. It appears you have refuted something Behe has not said. Congratulations.
At best, if this did wind up as disproving Behe, it is so small of an issue that it would only mean that Behe would need to re-formulate his argument. The difficulty goes up exponentially as more-and-more parts are needed for a system to work, so even if it IC was falsified in the small it does not mean that it would be falsified in the large.
I have to take Behe’s side on this — after ten years, this is the best they can do? Prove something false that Behe didn’t say, that was only moderately close to his smallest claim? And this is pumped through the media PR machine as if it is some breakthrough against Behe? Come on, give me a break. At least try to find something that _matches_ Behe’s definition of IC. It would be more interesting if one of Behe’s actual examples of IC from his book were targetted, but my guess is that noone wants to try.
Unfortunately, the local library doesn’t get Science until about a week afterwards, so I still haven’t read it. But, based on what I’m reading here, they are simply changing Behe’s definitions of IC and then claiming to have falsified them. The only thing I see in the response is a justification for having changed Behe’s definition.
Comment by Jonathan Bartlett — April 11, 2006 @ 10:55 pm
“At best, if this did wind up as disproving Behe, it is so small of an issue that it would only mean that Behe would need to re-formulate his argument.”
Yes, all he would have do is to move the goalposts. This is what creationists always do when their claims are shown to be false by scientists.
“Oh, OK, I was wrong, that wasn’t IC! But THIS is! Neener-neener!”
Comment by Don Baccus — April 11, 2006 @ 11:07 pm
Don–
Don’t be absurd. If you haven’t called a given system IC, you really don’t have anything to retract or any goal posts to move. The basic definition, constant since Darwin’s Black Box, was definitely not satisfied in this “rebuttal”.
But what’s more– supposing for the sake of the argument that something that actually fits the definition of IC was shown to come about by a mechanism that proved Behe wrong, there would be nothing low-class or “creationist” in reformulating the argument. That is simply how good science is done; we make hypotheses, we test them and make predictions; if the predictions come out right we’re happy, if they’re not, we go back to the drawing board and rework our hypothesis. It’s called the scientific method.
Comment by Freawaru — April 11, 2006 @ 11:52 pm
Andrew–
Thanks for that link to the address; it appears more to the point then the bit from the trial. I wouldn’t chalk it up as re–defining “irreducible complexity”, but it does seem he raised questions about systems similar to the ones being studied, and so calling them “piddling” now is probably not the best response.
At the same time, it looks as if very little new complexity or anything was generated; and no design theorists have an issue with simple mapping. But I see this is being discussed in some other comments above.
Comment by Freawaru — April 12, 2006 @ 12:03 am
‘”The basic definition, constant since Darwin’s Black Box, was definitely not satisfied in this “rebuttal”.’
Behe has given multiple definitions of IC.
And, no, that is not “simply how good science is done”. Good science is not done by sitting on your rear and simply changing your hypothesis whenever someone proves your hypothesis is false.
Comment by Don Baccus — April 12, 2006 @ 12:22 am
“It looks as if very little new complexity or anything was generated”
That’s rather the point. Not only is this not an example of IC, it’s evolutionary pathways are short, simple, and in no way improbable.
Comment by Don Baccus — April 12, 2006 @ 12:26 am
Don–
So…..
Comment by Freawaru — April 12, 2006 @ 12:28 am
“And, no, that is not “simply how good science is done”. Good science is not done by sitting on your rear and simply changing your hypothesis whenever someone proves your hypothesis is false.”
It is ludicrous to think that someone should get a theory formulated exactly correct the first time. That’s the _whole point_ of having journals and the like — to have others criticize your theory, and for you to go back and re-think and re-formulate your hypothesis.
Has the fact that the Big Bang needed to be re-worked several times indicate that it is false? No. The fact is that the Big Bang model had to go through numerous revisions to accomodate the data. And there’s nothing wrong with that. Don’t take a naive view of falsification. If that were the case, science would stop in its tracks.
But, as has been pointed out, we haven’t even arrived at that point yet, so all this talk is premature. But I did want to point out that even if the paper did what it was claiming to, it would not be a huge blow to ID — it would just mean that Behe would have to go back to the drawing board and re-think some of his concepts.
Comment by Jonathan Bartlett — April 12, 2006 @ 12:20 pm
Okay, so we can all agree that IC as it is presently formulated by Behe is no obstacle to evolution as currently understood by mainstream science?
(Jonathan Bartlett: I omit the adjective “Darwinian” because it’s not particularly accurate when describing the modern synthesis. It’s typically only creationists who use the adjective as a slur.)
Comment by Andrew — April 12, 2006 @ 2:15 pm
“Again, the binding site was _already there_.”
I repeat, the original organism and its different hormone-receptor systems lacked aldosterone and it’s receptor, including the aldosterone binding site. You are thinking of the ancestral state receptor, which was a different cortisol/DOC binding site.
To get from this situation to the aldosterone-MR receptor is an irreducible complex chicken and egg situation *according to Behe’s ‘multiple parts’ original IC definition*. (And furhermore according to Behe, the way this paper showed evolution did it, should not be happening.)
“Nowhere that I am aware of does Behe list increase or decrease in selectivity as being irreducibly complex.”
No, the point was that he tried to explain away the experiment by stating “Nothing new was produced in the experiment; rather, the pre-existing ability of the protein to bind several molecules was simply weakened.” which Musgrave show is false. The selectivity for aldosterone was produced. The sensitivity was perfectly as it should be.
“I have to take Behe’s side on this — after ten years, this is the best they can do?”
The point is that no ID research has proved IC, but that this paper, looking at an evolutionary process, happened by chance disprove one of Behe’s formulations. Thereby the concept of IC is refuted.
One of its problems is that there are several formulations, see Musgraves post at Panda’s Thumb that I linked to. If any ID proponent had been doing research for the last ten years they would have easily tested each and everyone of these formulations and perhaps selected one that survived initial tests.
The experiments they could have used should of course start with such simple systems as the current research looked at. To suggest that one should, or must, start with more complex systems is to not understand how science is done.
ID hasn’t done its basic homework, so the field is pushed further into obscurity. That the rest of us take home is that ID is moving its goalposts according to evolution progress, just as religion has always moved its goalposts according to science progress, and “that IC as it is presently formulated by Behe is no obstacle to evolution as currently understood by mainstream science” as Andrew remarked.
Comment by Torbjörn Larsson — April 12, 2006 @ 3:38 pm
“Okay, so we can all agree that IC as it is presently formulated by Behe is no obstacle to evolution as currently understood by mainstream science?”
It depends on what you mean by mainstream. IC is no obstacle to the same kind of processes by which ID is no obstacle to. However, the people who are screaming about ID are usually the ones (such as Leshner and Scott) which are committed to Darwinian paradigms. In fact, it is fairly clear that the existance of a structured search space (which is precisely what a non-Darwinian mechanism is) leads directly to the idea that the structuring was done by an intelligent agent. See:
http://www.4truth.net/site/apps/nl/content3.asp?c=hiKXLbPNLrF&b=784461&ct=1742245
According to the Berkeley website on evolution:
http://evolution.berkeley.edu/evosite/evo101/IIIC1aRandom.shtml
“In this respect, mutations are random—whether a particular mutation happens or not is generally unrelated to how useful that mutation would be.”
And this idea that organismal change is unrelated to organismal need has been exactly what has caused people to doubt teleological arguments. This is precisely the basis for Dawkins climb up Mount Improbable. Now that we know that organisms heavily rely on a mutation space that is structured to give beneficial, sensical mutations based on need, teleology is clearly implied, and the refutations of teleology is clearly refuted. This is why the “Darwinian” adjective is critical. This is why ID’s main aim has been Darwinian evolution. Other forms of evolution are perfectly inline, and actually support, ID ideas.
So, if you are willing to grant that the “mainstream evolution” supports ID, then I will concede that “mainstream evolution” is not a challenge to IC. If you at least continue to admit that non-Darwinian evolution is a major part of current evolutionary theory, then we can perhaps work together in getting objections to Darwinism heard in the classroom, since we can both agree that this is part of mainstream science. Currently, the Darwinian version is the only part of the story being told. If the Darwinian version is no longer mainstream, we can at least work together to get directed/assisted/whatever-you-want-to-call-it evolution into schoolbooks rather than the outdated Darwinian version.
“Okay, so we can all agree that IC as it is presently formulated by Behe is no obstacle to evolution as currently understood by mainstream science?”
This is patently false. I suggest you take the time to actually read the biological literature and see how Darwinian and neo-Darwinian is used. It is usually used exactly as ID’ers describe it, AND as a positive thing.
As a quick example, when deciding how to characterize certain genetic changes, this paper had a discussion about whether or not certain characterizations were neo-Darwinian, with the implication that they clearly preferred Darwinian explanations. In fact, in review articles discussing adaptive mutagenesis, the question of whether or not such processes can be accomodated by Darwinian paradigms are of great concern.
Also, in this paper, the author pleads with his audience that a given observation is compatible with Darwinism.
The fact is, that the “old guard” in science (the ones who are in charge of peer-review, for instance) are very strict Darwinists, and people have to bend over backwards to point out that what they are saying can be fit in with the Darwinian paradigm that it is very amusing. For example, see the abstract of Chance Favors the Prepared Genome:
“Rejecting entirely random genetic variation as the substrate of genome evolution is not a refutation, but rather provides a deeper understanding, of the theory of natural selection of Darwin and Wallace.”
It doesn’t matter how irrational it is, we _have_ to align ourselves with Darwin! If someone figures out that Darwin doesn’t matter, then Dawkin’s will lose his security blanket! Oh the horror!
Comment by Jonathan Bartlett — April 12, 2006 @ 3:45 pm
“I repeat, the original organism and its different hormone-receptor systems lacked aldosterone and it’s receptor, including the aldosterone binding site.”
False. (a) a binding site was there. (b) it already bound to aldosterone. Interestingly, the evolution of aldosterone (which I believe would _need_ a specialized binding site _before_ it would be useful [thus inhibitting both the selection of it and its receptor]) was not considered.
So, with all of your premises incorrect. No.
“If any ID proponent had been doing research for the last ten years they would have easily tested each and everyone of these formulations and perhaps selected one that survived initial tests.”
That would require major funding, and, last time I checked, there was not a large funding supply for ID. In addition, since this did not test IC, I don’t see why they would have tried this experiment.
“disprove _one_ of Behe’s formulations. Thereby the concept of IC is refuted.” [emphasis mine]
Even if this were true (which, as I pointed out above, it isn’t), it would only be a _partial_ refutation. If science proceeded on this basis, we would have rejected evolution long ago. Do you view the fact that Darwin was wrong about a lot of things as a total failure of his theory?
“The experiments they could have used should of course start with such simple systems as the current research looked at. To suggest that one should, or must, start with more complex systems is to not understand how science is done.”
I didn’t suggest that. What I did suggest was that taking the small bite is not the same as taking the big bite. They are taking the small bite (and, as many have shown, not even been successful at that) and _pretending_ to have taken the big bite. You do it in your own post. Had this not been publicized by a big PR push, and allowed discussion in the scientific community before calling the NY Times, noone would be calling foul. Behe would submit a response, clarifying his concepts and pointing out how the experiment did not meet them, and everybody would move on. But to treat such tiny claims as if they were huge, sweeping claims seems to be characteristic of the Darwin-pushers.
“That the rest of us take home is that ID is moving its goalposts according to evolution progress, just as religion has always moved its goalposts according to science progress”
That’s a huge case of the pot calling the kettle black. Again, I refer you to evolutionary theory and the Big Bang if you want to talk about “goalpost moving”. The fact is that all ideas must be reformulated with time. I don’t know of _any_ idea from _any_ arena of life which has not been. To say that this is restricted to ID and theology is simply being disingenuous.
Comment by Jonathan Bartlett — April 12, 2006 @ 4:02 pm
With all due respect, Jonathan, your latest series of posts are irrelevant to the issue at hand.
No one is saying that the Bridgham et al. paper refutes all of intelligent design. (That’s not even conceptually possible, since an omnipotent God can do whatever he wants. That’s why it’s fun to be omnipotent.)
So your efforts to resusciate ID based on “teleological explanations” and “structured search spaces” don’t even rise to the level of red herrings. They’re not talking about rejecting teleological explanations, so it’s just not relevant.
Rather, the paper takes a biological structure Behe previously characterized as irreducibly complex and shows an evolutionary pathway for the evolution of that structure.
Believe ID if you want. I’m sure I can’t convince you otherwise and I won’t bother trying. But can’t we at least agree on the obvious — that Behe looks like an idiot and a liar now for characterizing the Bridgham et al. paper as “piddling?”
Comment by Andrew — April 13, 2006 @ 5:23 pm
“No one is saying that the Bridgham et al. paper refutes all of intelligent design.”
Actually, the Science commentary _does_ in fact say that. If you think that this is an over-reaching statement, you should join me in criticism of that.
Here’s what the Adami perspective article says:
“such studies solidly refute all parts of the intelligent design argument.”
“So your efforts to resusciate ID”
What resuscitation? Nothing has been refuted.
“They’re not talking about rejecting teleological explanations, so it’s just not relevant.”
Perhaps you missed the part of the discussion where we talked about why Darwinism is an important part of the formulation.
“the paper takes a biological structure Behe previously characterized as irreducibly complex”
False. As pointed out many times above, Behe never characterized this as irreducibly complex.
“But can’t we at least agree on the obvious — that Behe looks like an idiot and a liar now for characterizing the Bridgham et al. paper as “piddling?””
No. Obviously you missed all of the conversation above and ignore all of the actual claims of both parties. Just because you _want_ this paper to be a win for Darwinism doesn’t change the fact that this paper does not in any way test irreducible complexity.
As Adami said, the information presented is “intrinsically interesting to biologists” — I agree with that completely. However, it is not of relevancy to the ID argument because it is not a system which Behe claims are irreducibly complex. How is this a difficult subject for people to grasp?
Comment by Jonathan Bartlett — April 13, 2006 @ 6:19 pm
Regardless of however many times you want to say “this isn’t a system [that] Behe claims [is] irreducibly complex,” it’s just not so. I and others have given you the link showing that protein binding sites are precisely the sorts of systems Behe has previously identified as IC.
The paranoid mindset of the creationist is amusing, if sad. I don’t “want this to be a win for Darwinism,” nor am I part of some vast Darwinist conspiracy to suppress contrary ideas, as you suggested earlier. (If that were so, Stephen Jay Gould would have been shot down in his infancy.) I just care when otherwise-intelligent people say demonstrably false things.
Comment by Andrew — April 14, 2006 @ 10:22 am
Andrew, your link showed that Behe thought it might be a system which displayed a real and highly interesting problem in biology, maybe even “irreducibility” in a sense, but it is still a stretch to turn that into a new definition of “irreducible complexity.” Really the only reasonable course of action is to stick with the actual definition.
I agree that it might not have been most accurate to call the results “piddling”; but then again, though they reflected a significant amount of work and looked at an interesting question, the results were extremely minor compared to the billing (think commentary in Science, NYT…)
Moreover, your comment:
is complete nonsense, since you are confusing ID with creationism (maybe you think they are the same, but we are discussing ID here, not creationism). Since ID theory says that there are certain features “best explained by an intelligence” and “which could not have come about by a natural cause”, a refutation is definitely within the realm of science.Comment by Freawaru — April 14, 2006 @ 3:50 pm
First of all, some opening comments, it is not well known, but Bill Dembski has been mildly critical of Michael Behe’s definition of IC. One will see Bill’s suggested alternative definition in his book, No Free Lunch.
Both myself and Mike Gene believe it is too difficult of a position to say that ALL IC systems can not evolve via Darwinian means, but IC does pose a very formidable and substantial barrier to Darwinian evolution, and that the abundance of certain IC systems would be sufficient to put design as legitmate alternative to naturalistic evolution.
That said, I think Behe is right to call this a piddling example and this paper is knocking down a strawman.
The examples Andrew cites with Behe deals with the exponentially difficult correlation of the number of amino acid positions to the number of trials required to achieve a protein with an appropriate characteristic. The examples that Behe and Snoke explored attempted to quantify the improbability when there were a number simultaneous changes needed and when there was low to negative selective advantage to the intermediates. 1 amino acid substitution is piddling, 2 is somewhat piddling, however because of exponenial growth, once one gets to 8 or more, it’s starts to no longer be piddling. Thus they helped to clarify the idea of what it means to have “SEVERAL well matched parts”, rather than just some vague notion of what SEVERAL means. Behe’s definition uses the phrase, “several well matched parts”.
In my book, 2 amino acid residues do not constitute “several well matched parts”, but that is what the supposed rebuttal dealt with. 2 amino acid components! A crude worst case search to find the combination is 1 out of 20^2 = 400, which is quite achievable.
Consider that I wanted to break a password that had only 2 alphabetic characters. That would be feasible, I need 26^2 trials. However, to imply that because one can resolve a password with 2 alphabetic characters that therefore a password of 10 characters would easily be achieveable is misleading. In the same way, the researchers have abused the idea of what “several” means in the definition of IC.
Further, Musgrave misrpresented Behe by implying Behe contradicted himself by rejecting the rebuttal because it dealt with a single protein, when the reason was the system lacked several well matched components. Of course, one might argue that 2 amino acids are “several well match parts”, but that is quite a stretch!
Behe might have argued his case better by comparing it to the proteins he and Snoke were describing, where the conception of the word SEVERAL was a quantified in greater detail.
Comment by Salvador T. Cordova — April 15, 2006 @ 2:51 am
So now we have established that IC is not really a good indicator of design, unless it is. Since IC and intelligent design itself are negative arguments and no positive hypotheses are presented, I would argue that IC just like ID is a scientifically vacuous concept based on our ignorance.
Yes, Dembski ‘corrected’ some of the flaws with Behe’s definitions and Behe never provided any evidence to support his claims about indirect routes. Now we find out that Darwinian routes and evolutionary routes (remember IC only focuses on Darwinian routes), can in fact generate IC. Which means that IC is not a reliable indicator of design after all. As such, IC has lost much of any relevance unless it can be saved by some positive hypotheses but that’s what ID seems to want to avoid at all cost (remember how Dembski considered such requests pathetic?)
Comment by PvM — April 26, 2006 @ 6:13 pm
Closing off old threads, I see this.
Jonathan,
“In fact, it is fairly clear that the existance of a structured search space (which is precisely what a non-Darwinian mechanism is) leads directly to the idea that the structuring was done by an intelligent agent.”
This entire comment is off-tangent. However, I will answer this specifically. A structured search space is not an evolutionary or nonevolutionary trait, it is the most usual search space. Natural slection for example looks at local gradient in fitness, there the fitness depends on the organisms possibility to answer to the current environment. Nothing unnatural is ‘doing the structuring’.
“False. (a) a binding site was there. (b) it already bound to aldosterone.”
There was no aldosterone produced in the original organism. The binding ability to aldosterone was incidential. This is how evolution proceeds, it can make use of available resources.
From the IC perspective neither aldosterone or its receptor existed. Thus, according to IC definition, this was an irreducible chicken-and-egg situation.
“In addition, since this did not test IC, I don’t see why they would have tried this experiment.”
My proposal was to test IC formulations. It is your concern how to do it.
” it would only be a _partial_ refutation.”
There is nothing in IC that distinguishes between 2 or more “irreducible components”. Thus the whole concept was falsified.
“Had this not been publicized by a big PR push, and allowed discussion in the scientific community before calling the NY Times, noone would be calling foul.”
So we can call everything of ID foul by its own PR practices.
“Again, I refer you to evolutionary theory and the Big Bang if you want to talk about “goalpost moving”.”
Evolutionary theories and Big Bang theories are much more elaborated than ID is. Details have changed and the theories have grown. However, IC refuses to meet experiments, see above. Since IC, even though it is only partly covering ID, is refuted all the assumed predictive parts of ID are falsified. (SCI is a non-starter.) It is a theological question whether ID is refuted or not; the assumed predictive parts are.
Freawaru,
“you are confusing ID with creationism”
ID trivially implies creationism, and vice versa.
Salvador,
“IC does pose a very formidable and substantial barrier to Darwinian evolution”
You have to prove this, it is unsubstantiated.
“1 amino acid substitution is piddling, 2 is somewhat piddling, however because of exponenial growth, once one gets to 8 or more, it’s starts to no longer be piddling.”
Your definition of IC makes it a bad indicator of design. Further it makes it unpredictive and unfalsifiable.
Comment by Torbjörn Larsson — May 11, 2006 @ 10:09 pm