That's gonna leave a mark: Jerry Coyne batters Behe

Uh, we can't see the article without registering and paying, so, uh...

will:

Huzzah for the cheapskates of the Internet who copy and paste with careless abandon!

:-)

If the "goal" of evolution is intelligent life, what can we conclude about a supposedly intelligent Designer whose greatest achievement is human beings?

I think Michael Behe aptly demonstrates that the process of evolution isn't particularly good at producing intelligence.

Behemothballed.

Will, if you want to read the whole thing try again - on the second shot I was offered a free view - but I still have not got it because you still have to register.

The list of pro-reality reviews and rebuttals has gotten pretty long and varied. What do you folks think of having an Edge of Evolution carnival? I've already gathered the URLs of posts which have appeared to date, so all that remains would be writing the summary prose to glue them all together.

Now don't all pile on me trying to tear me down on this, I've got a real question. I'm in your court, but confused. In the Dover case mentioned here, the judge apparently stated that "irreducible complexity was not evidence against evolution". Well, seems to me that it would be. Even Darwin said so. The thing is, of course, that irreducible complexity has never been proved. Even with the theists doing every thing they can to find something that would qualify. Furthermore, just because we don't currently know how a trait could be reduced in complexity and still useable doesn't mean it ISN'T reduceable. Is this what the judge meant? Whata ya think?

What I love most is Behe's fulminating against malaria and sickle-cell anemia. "You, virus, how dare you reproduce that way!? It's not helpful -- it's mere trench warfare!"

I can imagine IDers of two centuries ago claiming that God made the sun shine, because until 1938 we had no idea where all that energy came from. It was not until quantum mechanics arrived out of left field that the physicist Hans Bethe was able to surmise, correctly, that the sun is a giant fusion reactor, converting hydrogen atoms into helium and energy.

I wish Coyne had pointed out that evolutionary theory (and old-earth science in general) was once attacked on precisely these grounds by a genuinely great physicist. Lord Kelvin argued that the earth could not be as old as evolution and gradualist geology required, because no process known at the time could power the Sun for so long a period. That was a far more compelling argument than anything the creationID folks have produced in the last 100 years...and it was still wrong.

Evolution beat out its legitimate scientific competitors long, long ago.

I can imagine IDers of two centuries ago claiming that God made the sun shine, because until 1938 we had no idea where all that energy came from.

You'll be unhappy to learn that the Morris clan, in 2007, doesn't think that the sun is powered by fusion.

"Atheistic Theology?" Atheistic Theology?!?!" OK, that is just about the most assinine thing I've ever heard. How about those godless Christians...

Dahan,

Some biological systems are irreducibly complex. If they weren't, then geneticsists would never find strong phenotypes associated with null mutations.

A simple, yet excellent, discussion of why irreducible complexity (or interlocking complexity, as Herman Muller originally defined it from a scientific perspective over 50 years ago) is not evidence against evolution is on the Talk.Origins site. Check it out here: http://talkorigins.org/faqs/comdesc/ICsilly.html

Tex,

Got it. Thanks much. Nice site. Gosh I like learning. Lol!

Beautiful and devastating.

That's going into the permanent file.

Now don't all pile on me trying to tear me down on this, I've got a real question. I'm in your court, but confused. In the Dover case mentioned here, the judge apparently stated that "irreducible complexity was not evidence against evolution". Well, seems to me that it would be.

Irreducibly complex structures can evolve from simpler structures; just not directly. (Assuming, hah, a usable definition of "irreducibly complex"). You can have: simple starting point -> increasing complexity -> decreasing complexity along different path -> irreducibly complex endpoint. Or the old favourite: simple starting point -> long, involved, evolutionary history -> co-option of structures to new function -> refinement -> irreducibly complex endpoint. Of course you're only going to get irreducible complexity at a fairly abstract level, it's unlikely that all the fiddly details of a structure will be strictly necessary.
I suspect the ID dodge would be to claim that such endpoints aren't irreducibly complex. But that's a bait & switch; irreducible complexity is supposed to be a straightforward concept, not something that requires enumerating all possible evolutionary pathways to recognize.

Oh, and what the talkorigins site said too. I guess I should have looked there first.

I suppose if he loses his purchase on the "mutations are insufficient" crag, then he might try to stretch for a fingerjam in the "genetic drift isn't really random" crack. That will have a high comedy factor...

Scott Carrell has a review in Science as well. Did anyone else catch Behe's out-of-context quote of Carrell's Endless Forms Most Beautiful, near the end of The Edge of Evolution? I was reading the book in a bookstore (didn't want to support the man and his nine kids, ya know) so I don't have a copy to look it up again. I just remember it being made to say something like "We have to completely rethink evolution," which of course makes sense if you've read Endless Forms, but in Behe's book was ridiculously misleading.

Just a random thought (prompted by Coyne's mention of Lehigh's disclaimer)...I wonder how Behe would react to a prospective graduate student who wanted to conduct a lab study of ID under Behe's supervision. I'd love to see his response to something like that.

I wonder how Behe would react to a prospective graduate student who wanted to conduct a lab study of ID under Behe's supervision. I'd love to see his response to something like that.

A heart attack, perhaps?

Hii,
thanks for sharing this article, i enjoyed reading it..

Best Regards,
Eliena

I wonder how Behe would react to a prospective graduate student who wanted to conduct a lab study of ID under Behe's supervision.

I wonder if Behe has any students at present? He is a full professor, after all. He must get the occasional believing student who comes by and asks to help out with his world-famous work on irreducible complexity. How does he get rid of them without blowing the whole scam?

Or maybe he doesn't get rid of them. Maybe he's got a labful of students painstakingly printing out and ignoring immunology papers, hopeful that 5-6 years of this will net them a PhD. I would hope that he's canny enough to avoid another Bryan Leonard-style embarrassment, but after Dover, it doesn't seem likely....

To play devil's advocate: if we were to discover, say, a replicating organism with pistons and cylinders, fueled by an explosive reaction, should we then concede "intelligent design"?

Mr. Coyne's opening observations are very much similar to my first reaction, ie. that malaria and sickle cell are actually excellent illustrative examples of the mechanics of evolution.

here's my little screed on it:

http://tispaquin.blogspot.com/2007/06/wheels-off-wagon.html

Quoth Dahan:

Now don't all pile on me trying to tear me down on this, I've got a real question. I'm in your court, but confused. In the Dover case mentioned here, the judge apparently stated that "irreducible complexity was not evidence against evolution". Well, seems to me that it would be. Even Darwin said so.

Not quite. According to Behe, "irreducible complexity" means "all of the pieces are needed, intact and fully functional, in order for the system to function"; this doesn't say anything at all about how all the parts of the system got together in the first place. What Darwin said would be a problem for evolution is if there was some complex system that could not be accounted for as a series of modifications to some simpler system. Behe's IC is all about the system's current state; Darwin's example of what would be a problem for evolution is all about the system's historical process of development. Where Behe went wrong is his presumption that evolutionary stepwise modifications cannot generate IC -- and this presumption isn't just wrong, it's egregiously, glaringly wrong.
How could Behe have screwed up that badly? Easily, that's how.
Step one: Behe defines the term "direct Darwinian pathway" as a series of evolutionary steps, every one of which consists of a new part being added to the system. Okay, this is not particularly objectionable, in and of itself. Granted, it's true that, in addition to adding new parts to a system, evolution can also modify the existing parts of a system, or even remove existing parts of a system... but if you want to focus on that subset of evolutionary development which involves parts-addition and only parts-addition, more power to you. Hey, lots of science is done by focusing on specific points of interest and disregarding other stuff, right?
Step two: Behe demonstrates that it's not possible for an IC system to be the result of any "direct Darwinian pathway". Again, this is not particularly objectionable, in and of itself. Behe's "direct" Darwinian pathways are a rather limited subset of all Darwinian pathways, to be sure, but even so, it's at least mildly interesting to know about what sort of things cannot be done under certain sets of constraints.
Step three: Having demonstrated that "direct" Darwinian pathways cannot produce IC systems, Behe then goes on to conclude that no Darwinian pathways whatsoever can produce IC systems. And that is where Behe screwed up; he ignores all the other Darwinian pathways, the ones which involve at least one step that's not just adding one more part to the system.
Pathetic, isn't it?

To play devil's advocate: if we were to discover, say, a replicating organism with pistons and cylinders, fueled by an explosive reaction, should we then concede "intelligent design"?
Posted by: ken

If we were to find such a living organism on this planet, having no evolutionary relationship to any past or present organism, then one could likely conclude it was either an example of "intelligent design" or an alien life form which had evolved on a different planet. Neither would require supernatural intervention, as evolved intelligent beings could conceivably design a functioning life form.

.

It is also conceivable that "a replicating organism",

(M-W definition 1-a complex structure of interdependent and subordinate elements whose relations and properties are largely determined by their function in the whole) with pistons and cylinders, fueled by an explosive reaction)

would not be a living organism,

(M-W definition 2-an individual constituted to carry on the activities of life by means of organs separate in function but mutually dependent : a living being),

but a machine.

Let's try that last paragraph again:
.

It is also conceivable that "a replicating organism" with pistons and cylinders, fueled by an explosive reaction,

(M-W definition 1-a complex structure of interdependent and subordinate elements whose relations and properties are largely determined by their function in the whole)

would not be a living organism,

(M-W definition 2-an individual constituted to carry on the activities of life by means of organs separate in function but mutually dependent : a living being),

but a machine.

"Suppose a complex adaptation involves twenty parts, represented by twenty dice, each one showing a six. The adaptation is fueled by random mutation, represented by throwing the dice. Behe's way of getting this adaptation requires you to roll all twenty dice simultaneously, waiting until they all come up six... But now let us build the adaptation step by step, as evolutionary theory dictates. You start by rolling the first die, and keep rolling it until a six comes up. When it does, you keep that die "

Okay, I'm a tiny bit confused. I thought Behe's entire argument was that one six alone would have absolutely no advantage unless there were also nineteen other sixes. So I don't quite understand why we would "keep the six".

Don't get me wrong, I'm on your side and believe Behe would have to show that "a solitary six" *is* of no benefit which he never did (analogiously speaking). I just don't understand the argument that we will "keep the six" when our only criterion is "twenty sixes are good" but one, two, or even 19 sixes aren't any better than none.

What am I missing?

Woozy...you're correct about Behe's thinking. He thinks you've got to roll 20 sixes simultaneously. But Coyne says that's not how evolution works...you make one small improvement, keep it, and then move on to another improvement.

It's Behe's task to point out real examples where you've either got to roll 20 sixes, or nothing happens at all. And he fails.

Question: Behe figures that since you've got an error rate of 1/10^9 mistakes per generation, then the odds of two [simultaneous] mutations in the same gene would be 1/10^18. I'm wondering if that's empirically true...I mean, let's say your polymerase is a tad squirrely for some reason (translation error?)...then if it makes one error, odds are higher than average that it will make a second error. No?

It's interesting to note that the odds change radically if the mutation in question is one that allows the sperm to reach the egg more efficiently. In that case, you've got a pool of millions of sperm, each of which might carry beneficial mutation(s) that allow it to reach its target faster.

I'm still looking for the rabbit in the Pre-Cambrian.

Step two: Behe demonstrates that it's not possible for an IC system to be the result of any "direct Darwinian pathway".

I don't see how he has demonstrated even this smaller step. See Panda's thumb on this:

And in one of 200 simulations, the "irreducibly complex" and "Complex Specified" shape of the formal Steiner solution emerged. This is of interest because this is precisely the type of innovation that is supposedly precluded by Behe's "Irreducible Complexity" or Dembski's "Complex Specified Information."

Don't get me wrong, I'm on your side and believe Behe would have to show that "a solitary six" *is* of no benefit which he never did (analogiously speaking). I just don't understand the argument that we will "keep the six" when our only criterion is "twenty sixes are good" but one, two, or even 19 sixes aren't any better than none.

Simple. Behe sees everything in black and white, e.g., either you have a perfect eye or you are completely insensitive to light. In this world (one that is notable not part of reality) then you do need 20 6s to come up simultaneously. In the real world, where small advantages do count, (e.g., light sensitivity is an advantage even though it doesn't give you 4pi sr 20/20 color + IR + UV vision) it is only necessary for each mutation to provide a tiny advantage for it to be retained.

"Suppose a complex adaptation involves twenty parts, represented by twenty dice, each one showing a six. The adaptation is fueled by random mutation, represented by throwing the dice. Behe's way of getting this adaptation "requires you to roll all twenty dice simultaneously, waiting until they all come up six... But now let us build the adaptation step by step, as evolutionary theory dictates. You start by rolling the first die, and keep rolling it until a six comes up. When it does, you keep that die "

Okay, I'm a tiny bit confused. I thought Behe's entire argument was that one six alone would have absolutely no advantage unless there were also nineteen other sixes. So I don't quite understand why we would "keep the six".

Don't get me wrong, I'm on your side and believe Behe would have to show that "a solitary six" *is* of no benefit which he never did (analogiously speaking). I just don't understand the argument that we will "keep the six" when our only criterion is "twenty sixes are good" but one, two, or even 19 sixes aren't any better than none.

What am I missing?"

I think it is what Behe is missing. Rolling two sixes would not be that hard, but there would have to be an advantage at some time before the final row of twenty sixes. The thing is the advantage of 3 sixes and ten sixes might be something quite different from the advantage of 20 sixes. Things that initially evolved to give an advantage in one area are easily coopted by further mutation. The other thing is that when you talk about millions of bacteria or protozoa, improbable things happen all the time. A bit like winning a lottery. You wont win it, but someone will.

#18, "Scott Carrell" should be Sean B. Carroll.

#28, it should be understood as implied by the rest of the TNR article that each "6" would have to represent some minimal favorable adaptation, if Coyne didn't happen to mention that explicitly. (Didn't he?)

PZ, thanks for the link.

Anyone, is this representative of the caliber of TNR's writing? If it is, I may have to try to read it more often.

Thanks, PZ. Jerry Coyne is one of my favorite biological authors. Along with you and RD of course. His clarity of style does provide a complete knockdown of ID nonsense.
The trajedy, as he points out, is that unless someone is not properly skeptical of the sources one reads, one can be easily mislead. Behe is a blight on the Science of Biochemistry. Fortunately as Biochemists go, we have Larry Moran another good read.
Denis C

Hello, Denis! I trust you've been keeping yourself well since Darwin's?

Irreducible complexity...what am I missing?...I just happened to reread Mark Perakh's post (a few years old, I think) on just this question. He points out that 1) It would appear more probable that simplicity, rather than complexity, is the mark of design. Consider the complex shape of broken rocks, compared with the simple shape of a brick; 2) When humans design things (and isn't design by humans the basis for consideration of divine design, ever since Paley's watch?) they generally are designed so the entire thing does not fail if a single part fails. Behe simply hasn't made a good argument in support of design--just bad arguments against evolution.

What are you missing?

For starters, you seem to be assuming that the twenty-digit sequence won't exist at all unless it's composed of all sixes.

In real genomes, stretches of DNA can be redundantly duplicated, spliced differently, and repeated. There are also long stretches that don't seem to be obviously active and that are free to mutate as they please.

You're also not just rolling the dice once, but many times. If the sequence isn't seriously beneficial or harmful, it will be conserved for a while, although if we wait long enough it might be removed to increase efficiency. In that period of time, various things might happen to it, including generating stretches of sixes.

How many copies of the sequence (copies) exist? If there are a billion copies, or a trillion copies, that affects the odds. How long are we going to look at the population? If the sequence is copied once a day per organism on average, the length of time we observe the population will affect the odds that we'll see the twenty-digit stretch of sixes appear.

Finally, most fitness spaces aren't as narrow and specific as the one we've examined in this example.

That's just a taste of what you're missing.

About "irreducible complexity", I suggest looking at the Wikipedia article. One part of it is a very brief look at the "Forerunners" - three kinds of forerunners I find interesting: (1) Those who used the idea to argue against development of the embryo (as far back as the 1700s) (2) Those who pointed out that is was a consequence of evolution (as far back as the early 1900s) and (3) those who used it to argue for creationism (before Behe).

each "6" would have to represent some minimal favorable adaptation,

Not necessarily, it could also be a neutral mutation, or it could be a mutation that is negative on a "technicality", that is, a gene whose negative effect is not sufficient to be selected out of the population.

To play devil's advocate: if we were to discover, say, a replicating organism with pistons and cylinders, fueled by an explosive reaction, should we then concede "intelligent design"?

Personally, my working assumption would be fraud--the organism is a fake. But assuming fraud is ruled out, then yes, my next assumption would be intelligent design. The thing is that there are a large number of follow up questions: who are the designers? How do they think? Where does the fuel come from? How is it transported? Is the organism designed for self-maintenance? Maintenance by fellow organisms? Maintenance by something else? What functions is the organism suited to? How does it find the raw materials to replicate? Can it use raw materials in their natural state, or does it require refined starting materials?
If such an organism is not intelligently designed, then careful investigation, even starting from an assumption of intelligent design, is likely to find some evidence pointing to the true history of the design.

PZ: You got a free 6-month subscription to The New Republic Digital Edition with your subscription to Salon.com.

Thanks for the responses. The upshot I guess is that Behe is incorrect in his assumtion that there is no intermediary step toward "twenty sixes".

The thing is I was being both hypothetical and "giving the devil his more than he is due" in accepting (purely for the sake of argument) the criterion that anything other than twenty sixes are equally neutral. I figure if we accept this (bear in mind I'm viewing this only as a hypothetical exercise in probability) then if we somehow "mutate" one, two, or even ninteen sixes, those sixes could just as easily mutate out to a five or four before we mutate the final six.

I figure it would be a valid argument (but not of "intelligent design" which would be a terrible name for such a discovery, but of an as yet unknown process of mutation) but only if one can completely show that such an event is truly "irreducible complex" (which, to be honest, is a good name).

I've always suspected that human beings have a very naive intuiton about mathematics when it comes to large numbers. You find this in the insipid argument "dull athiest, could atoms random hurled form themselves to such a perfect world" or however that damned thing goes. Chaos Theory and Complexity theory I think show that surprisingly "perfect" things arise from simple randomness surprisingly fast. Unfortunately it's usually impossible to work backwords from the result to the initial state. (e.g. In the old computer game "Life" floaters and traffic lights nearly always pop up but it's impossible to see a floater and determine what it was before.)

Anyhow... I'm very curious as to how one does determine the "probability" of an organism. If one claims, even though I'd be sceptical, they found something "irreducibly complex" I'd be very interested in the math.

===
on another note, I really liked the point of artificial selection (being much faster than either natural selection or ID would be if it existed) being done by people *without* access to ID. I never thought about that.

I said: "...each '6' would have to represent some minimal favorable adaptation...."

Graculus said: "Not necessarily, it could also be a neutral mutation, or it could be a mutation that is negative on a 'technicality,' that is, a gene whose negative effect is not sufficient to be selected out of the population."

Well, OK, though to the extent mutations are neutral, or more so if they are even slightly negative, it will tend to elevate the chances that, in the time it would take to accumulate 20 "6"s, at least one of the 20 sites will mutate to a 1, 2, 3, 4, or 5.

Dahan@ #11,
additional to answer by Tex @ #12,
acknowledging the excellent post by Quentin Long @ #25
and the arguments for development by "scaffolding".

A modern system can be IC in the context in which it operates, without its being an obstacle to evolution. There are several ways in which this is possible. For example, our complicated blood-clotting cascade, or something like it, is necessary for organisms using a high pressure circulation, as has been discussed recently on one of the science blogs, but a system without some of its components is OK at lower pressures: so slight improvements in clotting made somewhat higher pressures possible, then further improvements, including occasionally adding components giving initially very little benefit (but capable of improvement), made even higher pressure possible, and so on.

Anton Mates @ #9
Kelvin's objection was actually that an old earth would have cooled to the extent that tectonic activity (not his terminology!) would have ceased. What drives the turnover in the earth's surface and thereby maintains long-term habitability is the decay of (primarily) uranium in the core, which depends on processes unknown at the time of Kelvin's objections, but discovered in his lifetime. The sun's energy supply was also seen as a problem in Kelvin's day, but there were potential (though ultimately inadequate) solutions, such as the continuing infall of distant materials: fusion of hydrogen to helium was of course the real answer.

Jim Roberts

Ah Lord Kelvin. Smart enough to formulate the 2nd Law of Thermodynamics, but not smart enough to see how it happens to be totally violated by... life. It's just too bad he didn't have as keen a mind as, say, convicted felon Kent Hovind.

/sarcasm

The sun's energy supply was also seen as a problem in Kelvin's day, but there were potential (though ultimately inadequate) solutions, such as the continuing infall of distant materials: fusion of hydrogen to helium was of course the real answer.

Kelvin argued, however, that these solutions would still not permit the sun to shine for evolutionary timescales. In fact, he specifically called out Darwin as demanding an unrealistic age for the earth.

"What then are we to think of such geological estimates as 300,000,000 years for the "denudation of the Weald"? Whether is it more probable that the physical conditions of the sun's matter differ 1,000 times more than dynamics compel us to suppose they differ from those of matter in our laboratories; or that a stormy sea, with possibly Channel tides of extreme violence, should encroach on a chalk cliff 1,000 times more rapidly than Mr. Darwin's estimate of one inch per century?....It seems, therefore, on the whole most probable that the sun has not illuminated the earth for 100,000,000 years, and almost certain that he has not done so for 500,000,000 years."

Well, OK, though to the extent mutations are neutral, or more so if they are even slightly negative, it will tend to elevate the chances that, in the time it would take to accumulate 20 "6"s, at least one of the 20 sites will mutate to a 1, 2, 3, 4, or 5.

Likely, but that doesn't prevent it from happening.

In the case of a copied gene, which is likely to be neutral, it actually speeds things up, because the copied gene, not being required for any function, is now free to go exploring. Some such genes could become positive factors if/when the population's environment changed, and make up for the slight drag at the beginning.

There's all sorts of ways and means in which mutations can eventually lead to a feature, it doesn't have to be a complete string of positives.

Sometimes copied genes have a positive effect, too, so there isn't even an initial drag.

Hey, is Behe still harping on the IC of the flagellum? I recently saw an article in Science that the current research is pointing that the F may have evolved from some molecular transport pores. Take that Behe!

The Michael Behe minivan, it's irreducibly complex my man!

http://thestubborncurmudgeon.blogspot.com

If anything, Behe's argument is irreducibly identical Zeon's Paradox of the Impossibility of Motion.

Jud @ 34:

TNR is run by a warmongering bigot who's not afraid to print counter-factual bullshit to promote his ideas about the way the world works. That should answer your implied question of if TNR is worth your money.

In case all the other links evaporate, Powell's has a nice, clean version on Review-a-Day.