New Work Documents the Evolution of Irreducibly Complex Structures

Posted 30 October 2004 by Dave Thomas

↗ The current version of this post is on the live site: https://pandasthumb.org/archives/2004/10/new-work-docume.html

I’ve recently obtained permission from Science and an author of the recent paper “Evolution of Coral Pigments Recreated,” to use the splendid figures in a popular discussion of this important new work. Permission to post these figures was granted only for the NMSR page, so I can’t post them here, but here’s a link to my new article, “New Work Documents the Evolution of Irreducibly Complex Structures.”

Here’s a snippet:

Recent work on the evolution of pigments in star corals, “Evolution of Coral Pigments Recreated,” by Juan A. Ugalde, Belinda S. W. Chang, and Mikhail V. Matz, (Science 2004 305: 1433 (9/3/2004), Copyright 2004 AAAS) shows conclusively that “irreducibly complex” structures not only can evolve, but that they have evolved. This should lay to rest the “Intelligent Design” assertion that this type of complexity is forbidden to natural evolution.

And Ugalde et. al.’s conclusion:

The more complex red color evolved from green through small incremental transitions (a stepwise accumulation of improvements), each identified in our experiments by ancestral gene reconstruction (Fig 1D). This mode of evolution has been anticipated since Darwin, but has only recently been demonstrated in computer simulation experiments (5, R. E. Lenski, C. Ofria, R. T. Pennock, C. Adami, Nature 423, 139 (2003), “The Evolutionary Origin of Complex Features” )

Continue reading “New Work Documents the Evolution of Irreducibly Complex Structures” (offsite at NMSR)

54 Comments

steve · 30 October 2004

Our study and understanding of ID tactics is sufficiently developed that we can formulate what the IDers will do in the form of laws. Call it Intelligent Design Theory Theory. Based on IDTT, which oddly enough has more scientific merit than IDT, I predict IDers will respond in the following ways:

1) Well, that system wasn't really IC in the first place
and
2) You didn't prove it, because there were computers involved / simulation isn't proof / the answer was preprogrammed / etc.

Joe McFaul · 30 October 2004

Perhaps a stupid question:

Has Michael Behe (anybody else) ever prepared a list of irreducibly complex structures other than the hints in Darwin's Black Box? I've got krebs cycle and flagella.. has he definitely identified anything else?

If IC is so useful, it woudl seem to be a simple matter for him to rattle off 30 or 40 irreducibly complex biological systems. Of course that would make his theory much more testable, wouldn' it?

Pete Dunkelberg · 30 October 2004

Thanks Dave, that's a neat bit of both science and irreducible complexity (IC). IC first became well known through Behe's 1996 book Darwin's Black Box. This appears to be a popular exposition of science, but there wasn't and still isn't any science behind it. Neither Behe nor anyone else has done any research showing that IC as such can't be produced by evolution. Why not? Because Behe's basic "IC can't evolve" argument doesn't even pass Biology 101. As you point out, blood clotting is one of Behe's favorite examples of IC. He tells the faithful that biologists who in fact know a lot about the evolution really don't. He lets on that he somehow refutes Ken Miller. But he only refers to the simplified account in Miller's book Finding Darwin's God, and is mum about the slightly more complex account at Miller's web site. Early steps toward the clotting cascade are known from lobsters and horseshoe crabs, which do not have a high pressure circulatory system and so are not solving the same problem. Turtles, amniotes like us, exhibit another stage in the evolution of clotting. The very greatest example of IC, according to the Discovery Institute (DI) from whence the nonsense flows, is the bacterial flagellum. But its evolvability in principle is clear enough. Dembski may not understand this, but Behe evidently got it. He could admit that he was wrong all along, or he could attempt the extreme step of claiming that even ordinary protein binding couldn't evolve. He tried the latter, finally publishing a model if not empirical research to back up what the said. Result:. But the DI admits nothing. By the way, the DI would have you believe that they invented IC and that it is a distinguishing feature of ID not found in older creationism. But take a look at this passage from Morris' famous book Scientific Creationism, second edition, p 59:

This issue can actually be attacked quantitatively, using simple principles of mathematical probability. The problem is simply whether a complex system, in which many components function unitedly together, and in which each component is uniquely necessary to the efficient functioning of the whole, could ever arise by random processes. Henry Morris, Scientific Creationism, second edition, p 59.
— Henry Morris

and compare

"By irreducibly complex I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning." Michael Behe, Darwin's Black Box, p. 39 (emphasis in original).
— Behe

Morris doesn't have ID's neologisms and does not claim a design detector as such. However, he argues that the complex systems he describes could not arise naturally. His treatment presages both Behe's rhetorical and Dembski's statistical "IC can't evolve" arguments. Check it at amazon.com.

steve · 30 October 2004

At least one way he's defined IC systems, there are trivially loads of them. Here's one way IC has been defined by Behe:

"By irreducibly complex I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning."

Now obviously you could find a thousand examples of such systems. That's not the problem. It's just a definition. Definitions aren't theories, they aren't hypotheses, they aren't much of anything by themselves. But in the next step, the IDiots say okay, things we've defined as IC, couldn't have evolved. This is a stupid statement, and can't be proved. There's no way to prove that any more rudimentary form of any such system, in any possible environment, is without benefit. Someone even thinking he can do so, tells us he's probably clueless. The number of ways that similar parts could be used in different systems, in different environments, is so unfathomably large that the idea of evaluating them is absurd. And so here, the IDiots fail, and are doomed to fail. Worse for them, a few times now they've pointed to something 'IC', and made the allegation that no precursor could have value, and been proven wrong.
So it doesn't matter whether they name 30,000 IC systems. They haven't show unevolvability, they can't show unevolvability, so they fail.

Well, they fail at creating a science. They succeed at convincing people they have.

Matt McIrvin · 30 October 2004

"I predict IDers will respond in the following ways:

1) Well, that system wasn't really IC in the first place"

Yeah... Reading the paper, I immediately imagined the IDer counterargument: "The initial green pigment was already irreducibly complex; you haven't shown how that evolved. The evolution to red just involved adding one more step; since you found a mechanism for it, that's not an irreducible increase in complexity, by definition."

Unfortunately, a real refutation would require pinning them down better as to what irreducible complexity is.

steve · 30 October 2004

I said "There's no way to prove that any more rudimentary form of any such system, in any possible environment, is without benefit." There's more than one way to interpret that sentence, given those ambiguous 'any's, so let me try to be clearer:

It's impossible to prove that no rudimentary form of such a system is without benefit in all possible environments.

a Creationist Troll, apparently · 31 October 2004

Naah. You're making far too big claims for a pretty small piece of research - what did that letter in Nature say about padding journals with overblown papers?

This is classic strawman stuff. You are misrepresenting the nature of "Irreducible Complexity", as far as I can tell. What you are saying, if I am reading your webpage right, is that because several steps are required to produce the red dye, and if any step isn't present the red dye won't be produced, then the reaction is irreducibly complex. However:
a) I don't think you demonstrate that there is actually a selective advantage for the coral having red dye (which would be required if natural selection is to be a driver of this process.
b) If the final stage of the process follows from a relatively small series of modifications from the initial stage, then even if it is "irreducibly complex", it is only trivially so, because you can derive the final stage from the initial stage. That is a long way from being able to demonstrate that PTA, Thrombin, Antihaemophiliac and Christmas could all independently evolve to make the blood-clotting cascade work.

Actually, it's not quite classic - because by saying "... and I bet that the refutation of ID'ers will be ..." you are dismissing a priori a valid counterargument i.e. this.

By the way, the list of IC examples in Darwin's Black Box are: bacterial flagella; blood clotting; cell transport mechanisms; the human immune system; biosynthesis of AMP.

a Creationist Troll, apparently · 31 October 2004

Naah. You're making far too big claims for a pretty small piece of research - what did that letter in Nature say about padding journals with overblown papers?

Actually, it's not quite classic - because by saying "... and I bet that the refutation of ID'ers will be ..." you are dismissing a priori a valid counterargument i.e. this.

By the way, the list of IC examples in Darwin's Black Box are: bacterial flagella; blood clotting; cell transport mechanisms; the human immune system; biosynthesis of AMP.

a Creationist Troll, apparently · 31 October 2004

Oh good grief!!!! Can't you get a friendly CompSci to find you some software that behaves to run this website on??? Yes, I did press "Refresh" - several times - before I reposted. Anyway, sorry for repeating myself.

PvM · 31 October 2004

This is classic strawman stuff. You are misrepresenting the nature of "Irreducible Complexity", as far as I can tell. What you are saying, if I am reading your webpage right, is that because several steps are required to produce the red dye, and if any step isn't present the red dye won't be produced, then the reaction is irreducibly complex. However: a) I don't think you demonstrate that there is actually a selective advantage for the coral having red dye (which would be required if natural selection is to be a driver of this process. b) If the final stage of the process follows from a relatively small series of modifications from the initial stage, then even if it is "irreducibly complex", it is only trivially so, because you can derive the final stage from the initial stage. That is a long way from being able to demonstrate that PTA, Thrombin, Antihaemophiliac and Christmas could all independently evolve to make the blood-clotting cascade work.
— Troll

THis is the classic and predicted response. Well it may be IC although you have not shown that there is a selective advantage (interesting strawman which ignores neutral pathways for instance) or that it may be an IC system but only trivially so. So is it or isn't it? Why moving the goalposts to PTA, Thrombin and so on? And who stated that these all independently evolved? Human imune system seems to have failed as a IC example already. THe bacterial flagellum is going that way with increased knowledge of homologies involved. Once again, ID has nothing to offer beyond well, in principle evolution could have done it but show us more detail. Fine, but lets not confuse our ignorance with intelligent design since ID once again has failed to present a positive hypothesis of design.

Wayne Francis · 31 October 2004

a) I don't think you demonstrate that there is actually a selective advantage for the coral having red dye (which would be required if natural selection is to be a driver of this process.
— aCTA

Where do you get the idea that for a trait to be successful it has to be of an advantage? While this may assist in the trait being propergated because the organism breeds more because of the trait it is NOT required.

By the way, the list of IC examples in Darwin's Black Box are: bacterial flagella; blood clotting; cell transport mechanisms; the human immune system; biosynthesis of AMP.
— aCTA

I've seen nice explanations, and even examples, of bacterial flagella, blood clotting, the human immune system. Pete Dunkelberg points toward the blood clotting here in this post. There are a few posts on this site that go into the bacterial flagella on the site and I'm pretty sure I've seen a post here about immune systems and the gambit of varieties that we see in nature. What most people would concider "transitional forms" of these systems from one solution to another. I swear that if rainbows went agianst the litteral word of the bible we would hear creationists deny they exsist and say there are no transitional colors from red to violet.

Pete Dunkelberg · 31 October 2004

Don't miss Evolving Immunity: A Response to Chapter 6 of Darwin's Black Box by Matt Inlay. There is probably material for a similarly detailed article on Evolving Coagualtion, but getting it all together is a job.

But keep your eye on the ball: The point of IC is that it is supposed to be in principle unevolvable. In reality, IC is a normal, hardly avoidable consequence of evolution. The rhetorical argument to the contrary is a smoke screen.

steve · 31 October 2004

The ICness of coagulation has been dispensed with, I seem to remember. The clotting system is really a collection of somewhat redundant subsystems. Redundant like this argument. Behe was backed down on it a while ago, and I'm sure someone here will post info about that.

steve · 31 October 2004

Creationist Troll said:

If the final stage of the process follows from a relatively small series of modifications from the initial stage, then even if it is "irreducibly complex", it is only trivially so, because you can derive the final stage from the initial stage.

Michael Behe said:

An irreducibly complex system cannot be produced directly (that is, by continuously improving the initial function, which continues to work by the same mechanism) by slight, successive modifications of a precursor system, because any precursor to an irreducibly complex system that is missing a part is by definition nonfunctional.

I knew these creationists didn't understand evolution, looks like they don't understand creationism either. It's a miracle they manage to get onto the internets.

Engineer-Poet · 1 November 2004

Our study and understanding of ID tactics is sufficiently developed that we can formulate what the IDers will do in the form of laws. Call it Intelligent Design Theory Theory.
— Steve

I suggest Intelligent Design Ideology Ontology Theory, or IDIOT.

steve · 1 November 2004

And here's some new work about the evolution of eyes.

http://www.embl.org/aboutus/news/press/2004/press28oct04.html

a Creationist Troll, apparently · 1 November 2004

Steve:

??? Saying that two proteins are similar says nothing at all about how evolution occurred! It simply says that the proteins are similar. It doesn't say how the protein might have come about in the first place - with any assessment of exactly what probabilistic barrier might have had to be overcome. And it doesn't say how any extra proteins might have come about that differentiate this "basic" eyesight from more complex.

There is 2x2 wood in the fence I built, and there is 2x2 wood in various places in my house, I don't doubt. But the fact that there is 2x2 wood in both places doesn't mean that a fence is a precursor to a house (or vice versa).

Is the example in the original "trivial"? Well, a classic substantial but basic example of IC would have two independent, different proteins, let's call them alphase and betase, with completely different structures. They would both be involved in catalysing a two-stage reaction from some molecule "Foo" to another molecule "Bar", via an intermediate "Wibble". "Wibble" would have no function for the organism. I have used those names deliberately (!) because they are obviously less connected than "A", "B" and "C". I may also be letting slip my own CompSci background!

Now, how does this relate to the paper cited? By the authors' own admission, the alphase and betase in this case are evidently very similar - and it is possible to suggest a series of changes that would allow mutations from one to the other - so they are not independent and different. And intermediate Wibble is not dissimilar to target Bar. It does lead to the organism having a different colour, but that's all that is claimed. That's what I mean by it being "trivial" - this is hardly "classical" IC, if there is such a thing.

I haven't looked at what has been written about the various examples of IC in Behe's book, nor the grounds upon which Behe apparently backed off his claims. However, I will say that I have been disappointed by the level of response to ID papers so far from the evolutionist community - they don't actually seem to do what they say on the tin. Sorry! But I will have a look when I get the chance, I promise.

Great White Wonder · 1 November 2004

There is 2x2 wood in the fence I built, and there is 2x2 wood in various places in my house, I don't doubt. But the fact that there is 2x2 wood in both places doesn't mean that a fence is a precursor to a house (or vice versa).

Wow. Now that you put it that way, Troll, I can see why ID is vastly superior to evolutionary biology. The wood in both places doesn't mean that a fence is a precursor to a house! Praise Jesus! Amen! I assume this analysis will make the cover of every major newspaper tomorrow as we now know beyond a mathematical doubt that a highly intelligent group of designers from outer space created life as we know it. So, Troll, from whom did you plagiarize this fantastic analysis? The Nobel Committee wants to know.

Dave Thomas · 1 November 2004

Well, I would have at least responded with a brief comment to our Creationist Troll that he's simply moving the goalposts, saying that the "Irreducible Complexity" of the coral pigmentation proteins "is only trivially so, because you can derive the final stage from the initial stage," all the while ignoring Behe's actual definition of IC, "wherein the removal of any one of the parts causes the system to effectively cease functioning."

But, Steve beat me to it! (Comment # 9786).

Thanks Steve!

Say, are you one of the >400 Steves for Evolution?

CHEERS, Dave Thomas

Engineer-Poet · 1 November 2004

(Neither did it occur to aCTa that the development of 2x4's for either houses or fences makes it much easier to build the other, even if that wasn't what the sawmill had in mind.)

steve · 1 November 2004

all the while ignoring Behe's actual definition of IC

Yeah, he ignored the crap out of that. He had to.

Say, are you one of the >400 Steves for Evolution? CHEERS, Dave Thomas

No, I won't have a Ph.D for another 5 years, if ever. But I do know one of them, Stephen Urquhart, because we both used to work for the same polymer physics group at NCSU. Everything else being equal, the number of physics Steves at NCSU is spooky.

steve · 1 November 2004

Oo, now there's 'trivial' IC and 'classical' IC. Add them to "Apparent" CSI, "Real" CSI, etc.

Apparently, creationist methodology is something like:

1) Make definition.
("anything with such and such is CSI")
2) Make assertion.
("CSI things can't have evolved")
3) Watch the above get obliterated.
("Here's a CSI thing that can evolve.")
4) Introduce Modifier
("Oh, that's just Apparent CSI. That's not Real CSI...")

RBH · 1 November 2004

Engineer-Poet wrote

(Neither did it occur to aCTa that the development of 2x4's for either houses or fences makes it much easier to build the other, even if that wasn't what the sawmill had in mind.)

Not to mention that the 2x2's (and 2x4's) in the house and those in the fence have common ancestry somewhere out there. Neither was "created" de novo out of nothingness. (Hm. Or is that what Troll is arguing? Hard to tell, actually.) RBH

Glenn Branch · 1 November 2004

Say, are you one of the >400 Steves for Evolution?
— Dave Thomas

Make that >500 Steves for Evolution, which includes someone named Tapani -- the Finnish cognate of Stephen. (Who knew?) The Stevometer is now at 512, but it looks better in binary: 1,000,000,000.

a Creationist Troll, apparently · 2 November 2004

OK. Ref: Behe's definition in comment 9290 above - does the system "cease functioning" if you don't do the last stage? No - it makes a different colour. Can you show that this has made it "cease functioning"? If not, you haven't shown that it is irreducibly complex, by Behe's definition. So I will go out on a limb (not being an expert, admittedly) and say that it is wishful thinking to call this irreducible complexity - and to say (in effect) that this proves that:

"irreducibly complex" structures not only can evolve, but that they have evolved. This should lay to rest the "Intelligent Design" assertion that this type of complexity is forbidden to natural evolution.

... and therefore that about wraps it up for ID - is absolutely absurd. You are drawing a mathematically precise line from an unwarranted assumption to a foregone conclusion.

Great White Wonder · 2 November 2004

OK. Ref: Behe's definition in comment 9290 above - does the system "cease functioning" if you don't do the last stage? No - it makes a different colour.

Oh, so the utility of Behe's "definition" of IC is highly dependent on how the "function" of the system is defined. Of course, we knew that already but it's useful to remind everyone just how bogus the "definition" is. And is it always the case that the creationist gets to define the function of the system? Is it also the creationist's duty to prove that the allegedly irreducibly complex system at issue has ONLY one function and that function has NEVER varied since the beginning of time?

steve · 2 November 2004

Awesome:

Posted by steve on October 30, 2004 05:48 PM I predict IDers will respond in the following ways: 1) Well, that system wasn't really IC in the first place

Posted by a Creationist Troll, apparently on November 2, 2004 03:12 PM does the system "cease functioning" if you don't do the last stage? No - it makes a different colour. Can you show that this has made it "cease functioning"? If not, you haven't shown that it is irreducibly complex, by Behe's definition.

I swear, there's nothing better than the Juxtaposed Quote.

Dave Thomas · 2 November 2004

A recap - the coral protein that catalyzes red colors requires three separate reactions. If any of these is altered, the function "Make Red" won't work.

Thus, it is "Irreducibly Complex" by Behe's own definition.

Behe ALSO states that any possible precursor won't have all the parts, won't perform the function, and therefore can't be selected for.

Behe's error is that he ignores the possibility that the precursor protein may be just fine for a different function.

All that Creationist Troll has shown us is that he, too, can be just as mistaken as Behe.

In other words, he can do a face-plant right along with the cream of the ID crop!

Cheers, Dave

Flint · 2 November 2004

It would help me if we could make a list of these errors. Dave Thomas contributes: 1)

he ignores the possibility that the precursor protein may be just fine for a different function.

Moving right along, I'm familiar with: 2) he presumes that examples of IC always evolve by adding parts, never by losing parts. The evolutionary path from reducibly complex to irreducibly complex can't be followed. 3) he presumes that if he can't imagine how something evolved, then it could not have evolved. This is argument by elimination. 4) his term "part" describes some level of aggregation suitable for his purposes, from individual protein molecules all the way up to entire complex systems. So "part" is like "kind", it means what it needs to mean under the circumstances. Are there more?

a Creationist Troll, apparently · 3 November 2004

Whatever. Just don't expect to be taken seriously if you think that that paper says anything substantial about whether IC systems can evolve. You haven't even suggested how the "Make Green" precursor might have appeared in the first place anyway. No. One more try. Let me use an analogy to try and demonstrate the scale of your heroic achievement. You have discovered that the phrase "Play together stay together" (PTST) can be derived by duplication of "Play together" (PT) and then a small amount of modification. Now PTST means something different from PT and ST (though it is related) - it is technically irreducibly complex. However: 1) You haven't demonstrated how PT might have come about on its own. 2) You are arguing that this proves that the following can also be derived from PT...

Theyve played soccer together for years most recently at Parkview High School My perspective is I just play with other soccer players and two of them happen to be my sisters Jada said Everybody else is like Thats so unique I see that it is and its a blessing that we get to play together, but weve been doing it all our lives If we didnt play together that would be strange to me That would be the phenomenon

(note that it includes duplication, interposition and so on). Also, although I know you are keen to simplify what Behe says to make it look ridiculous, Behe also makes clear that there is a graduation between "definitely evolvable" and "definitely IC" -

... the challenge will be to rigorously determine which systems were designed and which might have arisen by other mechanisms. To reach a conclusion of design will require the identification of the components of an interacting molecular system and the roles they play, as well as a determination that the system is not a composite of several separable syustems. To reach a strong presumption of nondesign will require the demonstration that a system is not irreducibly complex or does not have much specificity between its components

(my emphasis). So looks like, had you been familiar with what he said, you might have known that Behe had responded to this paper a few years before it was published.

PvM · 3 November 2004

Creationist troll accepts that IC systems can evolve from precursor parts and small modifications. Not surprisingly he thus focuses on show me how the precursor parts arose, missing the point that once again IC has shown itself to be a 'lil unreliable' in detecting design.
So how does ID explain the system... Oops I forgot, ID does not explain...

Pete Dunkelberg · 3 November 2004

IC as defined by Behe is a normal, virtually unavoidable consequence of evolution. How could you not sometimes get coadapted parts that together carry out a single function, as said function is perceived and specified by human observers? This is not mysterious (see earlier comments here) unless one is suffering from Demonic possession.

Troll is just trolling and playing semantics and moving the goal posts here and there to get attention. You know how it goes. A creationist will run all the way back to the Big Bang, never admitting all that has been conceded along the way.

steve · 3 November 2004

Just another Troll to show up for a few weeks, say the same things as countless evolution deniers before him, and eventually recede. Like Bob, David, FL,...etc...and apparently Pasquale, who's been scarce since he promised a mathematical disproof of Darwinism.

a Creationist Troll, apparently · 3 November 2004

steve :-) Probably. Well, I put on the bathroom wall something to back up what I was saying about evolutionist philosophers and WEASEL, by the way - I have to cross-promote or it will be ignored.

PvM, Pete. Yeah, yeah, whatever. But you don't win arguments by playing word games. I am not focusing on how the precursors arose - I'm just saying that all this paper is dealing with a very narrow series of steps in a process that in any case stretches the definition of irreducibly complex - and then saying that this solves all your problems. Perhaps you'd care to describe how my analogy is unrepresentative. Judging by the standard of argumentation that I've come across here, I imagine that most creationists get bored and wander off.

Great White Wonder · 3 November 2004

I am not focusing on how the precursors arose - I'm just saying that all this paper is dealing with a very narrow series of steps in a process that in any case stretches the definition of irreducibly complex - and then saying that this solves all your problems.

Okay, Troll, it's grammar time. When you say "I" and "I'm" in the above quote, are you referring to yourself? And when you say "this solves all your problems," what exactly is "this" and what are the "problems" you are referring to? At a minimum, a reasonable "standard of argumentation" requires that the parties involved use the English language correctly and consistently. The creationists I've seen around here don't wander off because they are "bored". They wander off because they are inarticulate morons with chips on their shoulders. After the chip is obliterated, all that's left is the inarticulate moron. Have you checked your shoulder lately?

Bob Maurus · 3 November 2004

Steve,

Just for the record, which Bob were you referring to?

steve · 3 November 2004

You, bitch!

;-)

No, your posts are good. I meant Bob Flynn, of course. And David is David Heddle, in case any other Davids out there are offended.

It's funny to me how many trolls have showed up to say the exact same things as their predecessor the previous week, and the week before that,...in general it doesn't look like they're very familiar with the general dialogue about evolution/creationism, any more than they're familiar with the scientific content of evolution.

timi · 4 November 2004

ACTA writes:
Behe also makes clear that there is a graduation between "definitely evolvable" and "definitely IC"

This is bass ackwards.
In Darwin's Black Box (DBB), ICness is defined via a "functional" criterion. Under this definition, which we call IC-Mark I (he's come out with other variants), a system can be determined to be "definitely IC" without regard to knowledge about its formational history. Thus "definitely evolvable" and "definitely IC" are not mutually exclusive descriptions of any particular biological system. In fact, Behe admits in DBB that some IC systems are evolvable.

This leads to the issue of figuring out how to distinguish evolvable IC systems from non-evolvable IC systems... Which kinda defeats the purpose of IC.

KeithB · 5 November 2004

"Early steps toward the clotting cascade are known from lobsters and horseshoe crabs, which do not have a high pressure circulatory system and so are not solving the same problem. Turtles, amniotes like us, exhibit another stage in the evolution of clotting."

I got to thinking about this while I was giving blood the other day. Does this mean that circulatory systems were developed very early in the Pre-Cambrian before the vertebrates and invertebrates split off? I kind of thought that there was little to no circulatory apparatus before the notochord came along.

Pete Dunkelberg · 5 November 2004

Squid, earthworms and us have what are called closed circulatory systems, with separate blood vessels. Most mollusks and arthropods have what are called open circulatory systems, but with contractile hearts. Blood and general body fluid are the same thing, better called hemolymph. Jellyfish and sponges have no circulatory systems in our sense, although even sponges keep water flowing through their bodies. Various other combinations are found. Search on 'circulatory' here for instance. Use google images for illustrations. Meet Limulus.

Bob Maurus · 6 November 2004

Thanks, Steve, that's a relief - I didn't think I was coming across as a closet creationist.

a Creationist Troll, apparently · 6 November 2004

GWW: So "you" (anybody else reading this) appear happy with my (the one I wrote, that is) analogy of what this paper (on coral pigments) achieves, then, since you (GWW) are more concerned about my grammar than defending the argument (of the paper).

I use the word "I" in the sense that being able to refute what I say wouldn't constitute demonstrating your position, since I am a very poor and unworthy opponent, one of the stupidest of men. Or possibly evil, but I'd rather not consider that.

timi: By "not being mutually exclusive" do you mean that "definitely evolvable does not imply definitely irreducibly complex"? You say, "Behe admits in DBB that some IC systems are evolvable". I don't think that's the case - perhaps you could say how you have drawn this conclusion. He does however say that not everything is either IC or definitely evolvable - so they aren't "mutually exclusive" in that sense.

David Wilson · 7 November 2004

In comment #9983:

You say, "Behe admits in DBB that some IC systems are evolvable". I don't think that's the case - perhaps you could say how you have drawn this conclusion.
— A Creationist Troll, apparently

I won't presume to speak for timi ( comment #9965), but on p.40 of Darwin's Black Box Behe did write

Even if a sytem is irreducibly complex (and thus cannot have been produced directly), however, one can not definitively rule out the possibility of an indirect, circuitous route [i.e. of evolution by natural selection]. As the complexity of an interacting system increases, though, the likelihood of such an indirect route drops precipitously.

Pete Dunkelberg · 7 November 2004

Let's keep our eyes on the ball. The important thing is not how much wiggle room does Behe appear to leave himself, but rather that the whole "IC can't evolve" story is bogus. IC is a normal, unavoidable consequence of evolution. Review: Behe introduces the impressive phrase irreducible complexity, or irreducible complex when used as an adjective, ('IC' stands for either) for one of the standard things that creationists insist can't be produced by evolution: co-adapted parts that jointly produce some result. He defines IC as follows:

"By irreducibly complex I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning." Michael Behe, Darwin's Black Box, p. 39 (emphasis in original).

Briefly, this amounts to "all parts required". Later, DI Fellows including Behe sometimes weaken it to "some parts required" (d'oh), but never mind that for now. Let's watch Behe express a great deal of creationism in a few sentences. While that big term irreducibly complex draws your attention, the little words part, system, and function surreptitiously do a lot of rhetorical work. These things are treated implicitly as given and static, which of course they are not. If your subject is physiology it may make good sense to focus on a certain function (as perceived by the observer) and the portion of the organism involved (system), and some subdivision or that portion into parts. These things are not likely to evolve very much during the course of a physiological investigation. But if your topic is evolution, thinking of functions, parts and systems as static is hopelessly myopic. It is also standard creationism: block change out of your mental picture and then conclude that change (evolution) is impossible. Quiz: As Behe's purpose in life is to defeat evolution and convince us that God is a great biochemist like him, let's stop at this point and try to think of a few ways in which IC might evolve. Think of as many as you can, and then you may peek. Next, Behe invents the division of evolutionary pathways into direct and indirect. He tries to define direct in such a way that direct evolution of IC is logically impossible (does he succeed?). Indirect is everything else. This brings us to what David Wilson quoted above [note that I omit the bracketed comment of DW's within his quote]:

Even if a system is irreducibly complex (and thus cannot have been produced directly), however, one can not definitively rule out the possibility of an indirect, circuitous route. As the complexity of an interacting system increases, though, the likelihood of such an indirect route drops precipitously. Behe, DBB p 40

If you want to understand creationist thinking, study that quote until the trick is perfectly clear to you. Suppose that Behe's definition of direct evolution does indeed make it logically impossible. Then everything is indirect. Are all of the ways you know for IC to evolve (you did the quiz didn't you?) so improbable that for all practical purposes IC just doesn't evolve? What's going on here? There are many indirect routes to a given point and no one of them is a priori likely. Similarly, no one bridge hand or cloud or snow flake is a priori likely. Does it follow that the whole category of snow is unlikely? Creationists, when thinking of evolution, think in that pattern over and over. Read the quote again and marvel at how smoothly Behe does it. Yet in reality, as the complexity of an interacting system increases, the likelihood that it evolved indirectly increases. Behe continues:

As the number of unexplained, irreducibly complex biological systems increases, our confidence that Darwin's criterion of failure has been met skyrockets toward the maximum that science allows.

Wow! But let's look at just the word 'unexplained'. What is it that is unexplained about IC systems? Their origin. Many pages before discussing any actual biological IC system, Behe has in his mind proven in a few paragraphs that no such thing can be a product of evolution. And the proof required no biology! Either he is a genius or else. There is an important next step in creationist thinking. As soon as something is explained, creationists including Behe automatically shift from arguing that it is in principle impossible to insisting that not all details are known, therefore it is still unexplained - in a different sense, but they seem to see it as the same thing. They can not see that their 'in principle' argument has been lost. Nor is it any use pointing out that one doesn't know every last detail about the historical origin of anything.

steve · 7 November 2004

Paraphrased Michael Behe: IC systems are unevolvable, except insofar as they are not.

(with apologies to Sir Arthur Stanley Eddington)

a Creationist Troll, apparently · 7 November 2004

OK. Take the digesting PCP example of irreducible complexity cited in the "quiz answer". There are three proteins involved, all of which were present in the original organism. We aren't talking about the appearance of a new protein to allow this bacteria to digest PCP's. All we are talking about is a change in regulation. This is a new metabolic pathway which is irreducibly complex, yes - but you haven't really solved the evolutionary problem - because you haven't said where the three proteins came from in the first place. Given that we are still learning about gene expression and regulation, it may be possible that this mode of expression was "there all along" - in the same way as Darwin's finch beaks change over time to deal with changing environment, the regulation of the expression of these genes may similarly change over time in response to changing environment.

In the context of mousetraps, you are at least saying that these three parts of this system didn't fit together before, but now they do. However, you haven't excluded the possibility that they weren't designed to fit together, which ties in with the fact that you haven't any inkling as to how they might have appeared in the first place. The odds against a protein with any functionality appearing from a random sequence of DNA is low - has any assessment been made of the odds of precursors of these three proteins being present "at random", and having sufficient functionality for mutations towards the currently specified forms of these proteins to have a selective advantage? Or (since in protein terms there are doubtless many ways to crack carbon rings) what is the minimum specification for a protein precursor to help with this reaction? Without a proper assessment you don't have a description of an evolving IC system; you have another IC system that you don't know how it might have evolved with the exception of the very last (and as usual, easiest) stage.

It's similar to the example given in the paper cited at the top of this thread.

At some stage, if evolution is to work, something somewhere has to produce new proteins. This is the part where creationists and ID'ers regard evolutionists' claims as truly incredible. If you can do that, then you can do anything you like. (Dangerous words! :-) )

timi · 7 November 2004

David Wilson wrote:

I won't presume to speak for timi ( comment #9965), but on p.40 of Darwin's Black Box Behe did write...

That's exactly the passage I had in mind. And Pete Dunkelberg nails it further. I'd add: On the "indirect pathway" idea, Behe presents a second definition of IC wherein the the degree of ICness is measured by the number of selectively neutral steps required during the evolution of an IC system. The idea with this second definition is to establish the improbability of a historical series of events. His thoughts are that systems requiring a large number of selectively neutral step are going to be highly improbable and hence unevolvable by natural mechanisms. The second definition of IC is thus a bit of a tautology: It defines a system based on its evolvability. To say, that an IC system (version 2) is unevolvable actually translates to: "an unevolvable system is unevolvable". Several problems with defining ICness by the number of selectively neutral steps: 1) Determing the ICness of a system under the second definition requires access to information that is mostly not available. You would have to know an awful lot about the actual history of the system before any such determination of ICness could be made. You would also need to be from at least a couple decades in the future because our current understanding of biochemistry is not sufficient to perform such calculations at this time. I am aware of no biological system that has been analyzed at this level and which met this criteria. 2) Two definitions of IC create massive semantic confusion: The original definition (IC = a system with multiple parts required for a particular function) is "ahistoric". It can be evaluated based on the state of the system as it exists today. In contrast, the "number of neutral steps" definition of IC requires historical knowledge of how such a system could arise. These are not synonymous definitions and it is not at all clear whether they have much overlap as applied to biology. We can determine that many systems probably meet the first definition, which tells us only that some systems come with nonremovable parts. However it is not known whether these same systems meet the second criteria, which could potentially indicate something about evolvability. What is particularly annoying is that many people fall into the trap of using these definitions as if they were interchangeable.

timi · 8 November 2004

One other addition: "Indirect" pathways are not necessarily the same as unselected pathways.

Wayne Francis · 9 November 2004

something somewhere has to produce new proteins.
— aCTa

DNA uses A adenosine C cytosine G guanine T thymine RNA uses A adenosine C cytosine G guanine U uracil uracil is equivalent to thymine. a base pair triplet forms an amino acid or peptide. Most amino acids have more then one base pair triplet that will function as that amino acid. There are 20 amino acids Alanine (Ala) GCU, GCC, GCA, GCG Arginine (Arg) CGU, CGC, CGA, CGG, AGA, AGG Asparagine (Asn) AAU, AAC Aspartic acid (Asp) GAU, GAC Cysteine (Cys) UGU, UGC Glutamic acid (Glu) GAA, GAG Glutamine (Gln) CAA, CAG Glycine (Gly) GGU, GGC, GGA, GGG Histidine (His) CAU, CAC Isoleucine (Ile) AUU, AUC, AUA Leucine (Leu) UUA, UUG Lysine (Lys) AAA, AAG Methionine (Met) AUG Proline (Pro) CCU, CCC, CCA, CCG Phenylalanine (Phe) UUU, UUC Serine (Ser) UCU, UCC, UCA, UCG, AGU, ACG Threonine (Thr) ACU, ACC, ACA, ACG Tryptophan (Trp) UGG Tyrosine (Tyr) UAU, UAC Valine (Val) GUU, GUC, GUA, GUG and Stop UAA, UAG, UGA start AUG (Note this is the same as Methionine amino acid) Proteins, or polypeptides, are made up of amino acids. This could be anywhere from 50 to 5,000 amino acids. Only about 4% of our DNA expressed as genes. The other 96% is used in things like regulation of theses genes. Now a random mutation in a base pair could have little to no effect. Such as with an alteration of Valine from GUU to GUC since both of these form the same amino acid. But what if the base pair that was affected was G in GUU to A. This would cause this amino acid to now be Isoleucine. What will this do in a gene? Maybe nothing maybe something maybe it will mess that gene up stopping it from functioning. What happens if the mutation isn't even in the genes but more likely in the 96% of the genetic code that is used for things like regulation of expression of these genes. In humans babies are born with a brain ~25% the size of his or her adult size. Via the expression of genes the child's brain grows to ~75% of its adult size in the first 2 years. We know that during this time of growth the brain is very pliable with neural pathways taking hold very easily, i.e. this is way children learn things faster then most adults. One of the big differences between chimps and humans is their brain development. At birth chimps and human babies develop very similarly. Intellectually humans take off and leave the chimps behind. Why is this? The genes that control the growth of the human brain stay switched on longer and the brain grows at a different pace. Did you know that a baby's brain grows at about 1.5 grams per day between 6months and 1 year? So alterations can be very minor but have major impacts. For some reason you seem to think for something to happen that a whole new polypeptide needs to be created. While this is possible, and probably more likely to happen because of some duplication then mutation event, this is not needed. Given that many mutations are lethal you don't have to worry about these mutations being passed on. By lethal I mean for a fetus. Most people don't even know about these situations. 15-20% of human pregnancies spontaneously abort. ~30% of these spontaneous aborts have abnormalities in their chromosomes and ~90% of fetuses with abnormalities in their chromosomes abort before birth. Humans tend to not be very flexible in chromosomal abnormalities. But this means about 10% of the abnormalities make it to a live birth. Hmmm strange this is a lot like what we see in the Equus group. There we have creationist saying they are all the same "Kind" and all derived from one pair that was on Noah's Ark. Yet we see drastic changes in their genetic code to include unique proteins not found in the other branches of that group. Something you claim is not possible. There are many things that mutations can cause. New proteins is just one that could occur. Development of a new organ, lets say one that senses gravity for as far as we know no animal actually has this ability, would not occur in one mutation. It would slowly evolve from many mutations. Perhaps giving the creature an advantage allowing it to more successfully breed thus passing down this trait. I don't need to know where the protein came from. All I have to do is understand how new proteins in a chromosome could evolve, which I apparently do but you can't seem to understand in your infinite wisdom, and then start to look at how we can trace what that new protein might do for said organism that it did not do, or did not do in the same manner, as it did before.

Wayne Francis · 9 November 2004

test to see if my last post really went thru

Dave Thomas · 9 November 2004

The Creationist Troll said

At some stage, if evolution is to work, something somewhere has to produce new proteins. This is the part where creationists and ID'ers regard evolutionists' claims as truly incredible. If you can do that, then you can do anything you like. (Dangerous words! :-) )

Yes indeed. Here's one splendid example, which even creationist Lee Spetner accepts as a "mutation producing new information." Spetner is quibbling about whether the mutation was random (probably was!, but that's another story. Thanks for the setup line, Mr. Troll! Dave Thomas

Wayne Francis · 11 November 2004

I see aCTa is strangle quiet. Maybe he's sick of say things like "So that doesn't prove anything" and making himself look stupid.

Great example Dave

Pete Dunkelberg · 13 November 2004

Hi "Troll" (aka aCTa), thank you for your follow-up in comment # 10000 to my comment # 9994. You agree with me more than perhaps you realize. Much of your comment is an implementation of my last paragraph: as soon as something is explained you shift to insisting that it isn't "really" explained because there is something else in the history of life that wasn't explained, even though that wasn't the topic. And you digress to other topics, even Darwin's finches. In saying that either genes or birds "change over time in response to changing environment", don't forget that mutation and selection and random drift are how the 'response' happens. And don't confuse short term evolution and long term adaptive speciation in Darwin's finches. Regarding the embarassingly easy evolability of IC (of the PCP IC system in particular) you say:

All we are talking about is a change in regulation. ... ... you have another IC system that you don't know how it might have evolved with the exception of the very last (and as usual, easiest) stage.

How could you forget that the whole IC 'problem' was the last stage; getting to an IC condition from the immediately prior non-IC condition. Since you don't have any evidence that some mutations are "easier" in some unspecified sense than a bunch of other unspecifed mutations, you in effect agree that the evolution of IC is easy. You ask where proteins come from, and whether they often have catalytic effect. Because of the physical-chemical properties of proteins, it is not unusual for a protein to have some catalytic effect on one or more orgainc reactions. But these might not be needed by the organism at the time. New proteins usually come from duplications of existing ones followed by mutations. These events include partial duplications, joining of two proteins into a single longer one, and adding part of one to another or to a duplicate of it. More is learned about different ways this happens all the time, for instance this laboratory study or this example from the wild. Do you want an entirely novel protein to just appear? That seems like too much to ask, and way off topic from the easy evolution of IC. But I'm in a generous mood ... have you heard of frameshift mutations? Recall that DNA is 'read' three bases (nucleotides) at a time, each three coding for an amino acid. An insertion or deletion of a single nucleotide in DNA upsets the 'reading frame'. As a result, a different sequence of amino acids is strung together into an entirely different protein. Voila! the orgainsm has a brand new protein to play with. Could it be useful? This is just what happened in the evolution of one of the ways that bacteria digest nylon oligomers. Isn't evolution wonderful? The evolution of IC is only one of its many marvels.

Steve · 13 November 2004

the sickle-cell mutation is in my mind the most amazing example of that.