Uncommon Descent Serving The Intelligent Design Community

Denis Alexander’s Strawman Just as Silly

Share
Facebook
Twitter
LinkedIn
Flipboard
Print
Email

In a comment to my last post T.lise picked up on another Darwinist strawman argument.  He quotes Denis Alexander saying:  “Many people impressed . . . of the huge improbabilities involved in biochemical systems coming into being ‘by chance’.  But what the reader might miss easily is that the calculations are based on the whole system self-assembling all in one go . . . But this is tilting at windmills.  No scientist believes that this is the way evolution works.” 

No ID theorist has ever argued that evolution is impossible because complex biochemical systems cannot self assemble “all in one go.”  This is an absurd caricature of the argument from irreducible complexity (IC).  

The basic logic of IC goes like this:  (1) By definition, evolution can work only in a stepwise fashion wherein each successive step is “selected for” because it has conferred a selective advantage on the organism.  (2) an irreducibly complex system is a system which if one part is removed all function ceases.  (3) by definition, therefore, an irreducibly complex system cannot be produced in a stepwise fashion.  (4) therefore evolution is not capable of producing an irreducibly complex system. 

Starting with this logic the ID proponent argues that certain systems are irreducibly complex and therefore could not have been produced by evolution.  The bacterial flagellum and the blood clotting cascade are classic examples of such systems. 

Again, no ID proponent argues that, for example, the bacterial flagellum could not be produced by evolutionary processes because it could not, like Athena from Zeus’s head, spring into being self assembled in one step.  

Now the theory of irreducible complexity is not particularly complicated in principle.  A smart guy like Denis Alexander surely knows that ID proponents don’t claim “System X could not have been produced by evolution unless it could be produced all in one go.”  So yet again we have an ID opponent apparently afraid to take on ID on its own terms.

Comments
Petrushka: Yes. When necessary. What's odd with that?gpuccio
December 22, 2011
December
12
Dec
22
22
2011
12:23 AM
12
12
23
AM
PDT
I think I must say that I really appreciate your comments: they are very pertinent, simple and clear.
Thanks, that's very kind. I too have enjoyed a civil and informed debate here.
As I have argue recently with Petrushka, I don’t believe that many basic new isolated biochemical functions that are naturally selectable are available in any definite biological context.
Forgive me, I haven't read your particular comments on the topic in the past. If you could link to anything pertinent I'd happily do so. In any case, I agree with you. It's interesting how we can both observe the same thing and interpret it so differently! I guess that's the heart of our paradigm differences, right there. What we see in biology is the repeated re-use of the same genomic components. Although this isn't an argument against design, per se, it is what we'd expect from natural processes when reaching entirely new function by searching sequence space is difficult. When new genes arise, they mostly arise through duplication, with neofunctionalisation coming through subsequent modification. What this seems to indicate is that neofunctionalisation doesn't require the wholesale, bottom up construction of amino acid chains but alteration of a few bases within. This does not mean that moving between different functions is 'easy' but it does make it less of a problem than is sometimes perceived. One thing that we can't do is reconstruct an ancestral sequence and then count the number of steps in between, assume each was necessary, and then calculate the probability of the protein having arisen via evolution. Gauger and Axe did something similar - although even less realistic - in their recent paper. This is post-specifying that a particular protein-coding sequence *had to* make a particular transition to a particular function via a particular series of steps. Exactly, as Petrushka has said, this is painting a target where an arrow has struck and is - of course - not how evolution is predicted to proceed. When parentless genes arise - and you gave a nice example from the Li et al. paper - the exons appear to largely be the result of insertions from other ORFs like transposable elements. Again, the interpretation that appears most simple to me is a naturalistic one. I don't see this pointing to a designer. Of course a theoretical designer could do whatever he/she/it wants, but the lack of creativity from such extensive re-use does not match up with what we'd expect based on our own designs, where innovation, new materials, new ways of doing things are always sought. This has at least some bearing on a design inference, because it is human design that is used as a standard to detect the design of an unknown designer.
And please, consider that the whole known proteome (at present), the result of 4 billion years of evolution, consists of about 2000 protein superfamilies.
Potentially pointing to the limits of evolution rather than the number of potential proteins left 'undiscovered'.
I would like to say that exons often correspond to functional units, and therefore exon shuffling, or domain shuffling, or alternative splicing, often correspon “simply” to a different assemblage of functional units, that is a very basic modality of modular design.
Yes, fair enough. But this also points to the fashion in which evolution is postulated to work. The endless recycling of past chunks of things - ancient things often. Such things are expected to be rare to the point of only occasionally arising, which is what we see. As an opinion, I'd have thought a designer with good knowledge of function in sequence would perhaps generate lots of novel things without this long history to them, but we don't really see that happening. I hope these brief comments can help give an indication of where my thoughts lie on the matter - briefly jotted down in between other work :)paulmc
December 21, 2011
December
12
Dec
21
21
2011
05:24 PM
5
05
24
PM
PDT
And please, consider that the whole known proteome (at present), the result of 4 billion years of evolution, consists of about 2000 protein superfamilies.
I find that argument rather odd. You are basically saying the Designer drops by every two million years or so and deposits a new protein domain. Sometimes more often, sometimes less, but interspersed throughout the history of life.Petrushka
December 21, 2011
December
12
Dec
21
21
2011
01:34 PM
1
01
34
PM
PDT
paulmc: I think I must say that I really appreciate your comments: they are very pertinent, simple and clear. That said, a couple of comments on your last post: Do you really think we can calculate this? Within an undirected evolutionary paradigm any novel, unexpected, function could be beneficial. Obviously, it’s a small ratio whatever the number may be, but I doubt we could ever accurately characterise it. Yes, I think so. I believe that the Durston method is a simple and effective way to approximate the target space for a protein family. If you have read other posts of mine, you may know that my main argument is about the emergence of basic protein domains. Let's say the 2000 protein superfamilies in SCOP. The argument that "any novel, unexpected, function could be beneficial" is IMO weak and not convincing. As I have argue recently with Petrushka, I don't believe that many basic new isolated biochemical functions that are naturally selectable are available in any definite biological context. The main limiting factor is the existing complexity of the system, that would allow only complex and integrated subsystems to work and be useful to the point that they give a reproductive advantage. The second limiting factor is that most useful subsystem would be irreducible complex. The third point is that, even if some finite and not too big number of single new protein families were potentially selectable in a specific situation, thir probabilities should be only added one to the other. So, if there were 1000 potential new protein superfamilies naturally selectable in a prokaryote, that would increase the target space only of three orders of magnitude (about 10 bits). In a scenario where the value of complexity are of the order of hundreds of bits, that would scrcely make any difference. And please, consider that the whole known proteome (at present), the result of 4 billion years of evolution, consists of about 2000 protein superfamilies. For all these reasons, I think it is extremely reasonable to accept the point of view of ID about protein domain probabilities. Yet, if the numbers were as dire as IDists suggest, what would the odds be that such a crude mechanism as exon skipping or another method of alternative splicing could produce a novel, functional protein? First of all, we should also consider the probabilistic resources and functional space and search space of those other nechanisms of variation. I have never tried, and I believe that at the present state of knowledge it is better to stick to the emergence of basic protein domains, which is the simplest ID scenario. But only a detailed analysis of those mechanisms could tell if they are really random and crude, or guided by design. You must remember that the only real method to distinguish between true random events and designed events a posteriori is the ID theory. Darwinists have renounced to try to do that, but that is not an acceptable scientific position, and is indeed very convenient for a theory that is based on imagined results of randomness, never verified or quantitavely assessed in a correct statistical context. Even so, I would like to say that exons often correspond to functional units, and therefore exon shuffling, or domain shuffling, or alternative splicing, often correspon "simply" to a different assemblage of functional units, that is a very basic modality of modular design. Also, the degree of interspecies variation in homologous proteins suggests a lower degree of specificity than is sometimes acknowledged. When the function is preserved, and so is the structure, sequence variation can be reasonably explained as the result of neutral or quasi neutral mutations. See the theory of the big bang of protein space. The Durston method takes that into account.gpuccio
December 21, 2011
December
12
Dec
21
21
2011
01:09 PM
1
01
09
PM
PDT
Also, the degree of interspecies variation in homologous proteins suggests a lower degree of specificity than is sometimes acknowledged.
The simple fact is that ID advocates have no theory at all that would shed light on the probability of sequences evolving. No way of evaluation sequences that would tell you how close they are to being functional, or even what function they might have. Even if you could determine whether a sequence codes for a protein, there is no theory that would help determine whether the sequence has utility for an organism (or whether it might induce disease). And of course, the existence of regulatory networks just compounds the problem.Petrushka
December 21, 2011
December
12
Dec
21
21
2011
12:12 PM
12
12
12
PM
PDT
a) The ratio of the target space to the search space
Do you really think we can calculate this? Within an undirected evolutionary paradigm any novel, unexpected, function could be beneficial. Obviously, it's a small ratio whatever the number may be, but I doubt we could ever accurately characterise it. Yet, if the numbers were as dire as IDists suggest, what would the odds be that such a crude mechanism as exon skipping or another method of alternative splicing could produce a novel, functional protein? Also, the degree of interspecies variation in homologous proteins suggests a lower degree of specificity than is sometimes acknowledged.
b) The probabilistic resources (number of attempts available, usually depending on population number and reproductive rate and time available).
Yes, this is important. But we need to be careful that we are not pre-specifying something that need not have happened.paulmc
December 21, 2011
December
12
Dec
21
21
2011
12:02 PM
12
12
02
PM
PDT
Drift leading to adaptation was observed in the Lenski experiment. A functional, near neutral intermediate sequence was confirmed by Thornton. Since it took Lenski about 20 years of painstaking research to confirm a three step adaptation, it is not realistic to expect something as complex as a flagellum to evolve in a laboratory experiment. But it is more likely than a designer sighting.Petrushka
December 21, 2011
December
12
Dec
21
21
2011
09:35 AM
9
09
35
AM
PDT
paulmc: Just to be clear, I am not denying that drift is a form of variation. But it is a form of random variation. It adds nothing to the more general concept of RV, which obviously is not limited to single point mutations. RV includes a vast range of variation mechanisms, including drift. But all of them are random. When we are evaluating the probabilty of finding a functional target in the search space by a random walk from an unrelated state, only two things count: a) The ratio of the target space to the search space b) The probabilistic resources (number of attempts available, usually depending on population number and reproductive rate and time available). Given those variables, what kind of RV we are considering does not change anything. Each new state attempted by variation is anyway a random state in a random walk. On the contrary, positive natural selection, if and when it occurs, determines an expansion of a specific form of variation, and a multiplication of the probabilistic resources for that trait. If the selectable trait is also an intermediate step to some more complex function, then the search for that function is no more a purely random walk, and must be reevaluated. That's why RV + NS, if and when applicable, is a powerful algorithm, while RV alone, whatever its form, is just a random walk.gpuccio
December 21, 2011
December
12
Dec
21
21
2011
02:47 AM
2
02
47
AM
PDT
paulmc (4.1.1.3.8): I think we are getting confused by terminology. You are using “modern synthesis” to mean “the assemblage of different evolutionary mechanisms that most modern evolutionary theorists employ”, whereas when most ID people speak of “modern synthesis”, they mean “The Modern Synthesis”, i.e., the “neo-Darwinism” of Mayr, Dobzhansky, and Gaylord Simpson, which was the ruling view in evolutionary theory for decades. Its formula was random mutation plus natural selection. You are right, however, to say that ID people also criticize some other proposed evolutionary mechanisms which postdate The Modern Synthesis. The reason for that is that none of those mechanisms, e.g., “drift”, seem capable of actually building any new biological form. For that matter, not even all critics of neo-Darwinism have anything much better to offer. Lynn Margulis’s mechanism of combining genomes has some application at the level of microorganisms, but beyond that level it seems a chancy and erratic mechanism to generate macroevolutionary change. What is needed is something that transcends freak mutations, freak genome combinations, and the endless shufflings of population genetics. What is needed is something that can explain apparent evolutionary tendencies toward useful form. In various ways, Conway Morris, Denton, Shapiro, Sternberg and others are all working towards an explanation of novel form that does not depend, ultimately, on blind searches. But of course the old guard, whom you apparently represent — whether neo-Darwinian or not — are going to defend the fortress on which their careers were built, until they leave the scene. Thus, no new ideas are going to come from Dawkins, Coyne, Lewontin, Eugenie Scott, Ken Miller, Francis Collins, etc. The future of evolutionary theory will not be dominated by population genetics, as it was in most of the 20th century. It will be increasingly shaped by input from fields traditionally distinct from biology: physics, engineering, computer science, information theory, pure mathematics. It is understandable that the population geneticists are a bit irritated that “evolution” will no longer be their private playground, where they don’t have to answer to anyone. They probably think that all these new “immigrants” to biology will lower the property values. But for ID people the future of biology is looking brighter than it has looked at any time since 1859, provided that the younger voices are not strangled in the crib by their jealous elders. T.Timaeus
December 21, 2011
December
12
Dec
21
21
2011
02:01 AM
2
02
01
AM
PDT
gpuccio, just to be clear I am not denying a place for positive selection. But removing genetic drift from the equation only leads us to the unfortunate strawman of classical darwinism, rather than the more complex, but also more realistic, contemporary version of evolutionary theory. If I understand you, you are arguing that drift is essentially unhelpful because it is difficult to see how it would assist in producing complex, specified information. In fact, drift is rather effective at maintaining the variation that may later play important evolutionary roles.paulmc
December 21, 2011
December
12
Dec
21
21
2011
01:05 AM
1
01
05
AM
PDT
paulmc: I would just like to mention that genetic drift and neutral evolution, as I have argued many times, are of no help in overcoming the probability barriers outlined by ID theory. That's why ID concentrates most on the mechanism of RV + NS: at least, that is a mechanism that could potentially work, if its false assumptions were true.gpuccio
December 20, 2011
December
12
Dec
20
20
2011
11:58 PM
11
11
58
PM
PDT
ID proponents don’t refer to all evolutionary biologists as Darwinists, at least, the more careful ID proponents don’t.
From the glossary here at UD: Darwinism: When ID proponents on this site use the term “Darwinism,” they are referring to Neo-Darwinism, also called the modern evolutionary synthesis or Neo-Darwinian evolution (“NDE”). So, the policy of this website is to refer to the bulk of mainstream evolutionary biologists as "Darwinists" regardless of how poor a description that actually is. The modern synthesis of course encompasses neutral evolution - to subscribe to the modern synthesis is not by any means to claim that every step in an evolutionary process is the result of +ve selection.paulmc
December 20, 2011
December
12
Dec
20
20
2011
08:25 PM
8
08
25
PM
PDT
So, they didn't mention it at all, did they? Paul, do you think representation-protocol-effect is an irreducibly complex system to transfer mutable genetic information?Upright BiPed
December 20, 2011
December
12
Dec
20
20
2011
06:43 PM
6
06
43
PM
PDT
paulmc: ID proponents don't refer to all evolutionary biologists as Darwinists, at least, the more careful ID proponents don't. Indeed, this site, long before other sites (atheist or TE), tried to get the world interested in non-Darwinian formulations of evolution. I first heard about the work of Shapiro and the Altenberg group here. And many on this site have promoted the work of Michael Denton, another non-Darwinian evolutionist, and Sternberg, who is still another. ID is not opposed to "evolution" or "evolutionary biology" as such. It is opposed only to such forms of evolutionary theory (Darwinian or otherwise) which rely entirely upon unguided mechanisms to produce novel biological form. By the way, there are still plenty of ultra-Darwinists around, including Dawkins, Ken Miller, and Eugenie Scott. You appear to agree with me that the classical Darwinian account must be abandoned. You would perhaps also agree with me (against Petrushka) that Denton has abandoned it. Where we disagree is that you merely supplement the Darwinian mechanisms by other chancey mechanisms, whereas I, with other ID people, deny that the sum total of all the mechanisms propounded by Mayr, Kimura, etc. can produce radical biological novelty. T.Timaeus
December 20, 2011
December
12
Dec
20
20
2011
06:41 PM
6
06
41
PM
PDT
paulmc and Petrushka- when has genetic drift ever been observed or hypothesized to construct new, useful multi-part systems? BTW Darwin used natural selection as a designer replacement/ designer mimic, not for all change. That is Darwin used it to explain the things that drift cannot. To explain complex proteiin machinery by referencing drift you may as well use sheer dumb luck.Joe
December 20, 2011
December
12
Dec
20
20
2011
06:27 PM
6
06
27
PM
PDT
He was arguing for the implausibility of naturalistic evolutionary mechanisms to account for some complex features of biology. In this context, genetic drift is fairly important.paulmc
December 20, 2011
December
12
Dec
20
20
2011
06:25 PM
6
06
25
PM
PDT
Timbo, If by sounding like a YEC you mean someone who goes by what the evidence says by applying our knowledge of cause and effect relationships, then yup, you got me.Joe
December 20, 2011
December
12
Dec
20
20
2011
06:21 PM
6
06
21
PM
PDT
He wasn't arguing against evolution either. How many times did Jerry Coyne mention representations and protocols in Why Evolution is True? How many times did Ken Miller mention semiosis when he wrote Only a Theory? How many times did Richard Dawkins mention the observable dynamics of symbol systems in The Blind Watchmaker? How many time did any of them mention anything at all about the immaterial relationships required for recorded information to even exist and be transferred?Upright BiPed
December 20, 2011
December
12
Dec
20
20
2011
06:15 PM
6
06
15
PM
PDT
In Darwin's Black Box Behe includes genetic drift once in a list of evolutionary mechanisms, without any discussion of its consequences. In The Edge of Evolution he mentions genetic drift not at all.paulmc
December 20, 2011
December
12
Dec
20
20
2011
05:38 PM
5
05
38
PM
PDT
You'd think that somewhere in a four hundred page book Behe would have mentioned neutral drift. Oh well, at least we can be certain that he mentioned ERVs in the list of evidence supporting common descent.Petrushka
December 20, 2011
December
12
Dec
20
20
2011
05:08 PM
5
05
08
PM
PDT
Oh, I agree, it's fun mostly. I was just struck by the hubris of post 4.1.1.Timbo
December 20, 2011
December
12
Dec
20
20
2011
04:12 PM
4
04
12
PM
PDT
[Petrushka's] grasp of history of evolutionary theory, and of the crucial assumptions of the theory, appears to be poor. If the individual steps are not selectable, the classical Darwinian approach must be abandoned.
The classical Darwinian approach where all change must be the result of positive selection was in vogue in the 1930s based on Fisher's (1930) mathematical arguments, but even then was not universally accepted (Sewall Wright, for instance argued for the importance of drift concurrently). Since the 1960s, this type of classical ultradarwinism has been abandoned at the molecular level (e.g. Kimura 1968). This is one of the reasons why it frustrating that IDists insist on referring to evolutionary biologists as Darwinists, a ridiculous and misleading title. At the morphological level, there is certainly evidence of positive selection for traits in some instances, but there is little reason to exclude large amounts of nearly neutral change by genetic drift resulting from comparable change at the molecular level. There simply are no robust, modern predictions in evolutionary biology that every change must be the result of positive selection, nor theory to form the basis of such predictions.paulmc
December 20, 2011
December
12
Dec
20
20
2011
04:09 PM
4
04
09
PM
PDT
Oh, that’s right, the whole mainstream scientific community are brainwashed/engaging in an intellectual conspiracy. Exactly. Well, at least you can “prevail” here… Well, at least I can say what I think here. And you too. And whoever prevails, prevails. It's not bad.gpuccio
December 20, 2011
December
12
Dec
20
20
2011
04:05 PM
4
04
05
PM
PDT
You sound like a YECTimbo
December 20, 2011
December
12
Dec
20
20
2011
02:43 PM
2
02
43
PM
PDT
Carry out research as opposed to your position's propensity of throwing unresearchable eons of time around to solve all problems? Who needs research when you have the "Father Time" trump card? All that is left is "cleanup on aisle 7"....Joe
December 20, 2011
December
12
Dec
20
20
2011
02:06 PM
2
02
06
PM
PDT
If the individual steps are not selectable, the classical Darwinian approach must be abandoned.
Interesting that you accuse me of not grasping the history of evolutionary history.Petrushka
December 20, 2011
December
12
Dec
20
20
2011
01:49 PM
1
01
49
PM
PDT
Well, perhaps it is time to venture out of the confines of an ID friendly blog, carry out some research, and present your findings to the scientific community. Maybe write a paper or two for publication. Oh, that's right, the whole mainstream scientific community are brainwashed/engaging in an intellectual conspiracy. Well, at least you can "prevail" here...Timbo
December 20, 2011
December
12
Dec
20
20
2011
01:36 PM
1
01
36
PM
PDT
Good work, Eric and gpuccio. Petrushka may or may not have some understanding of some of the technical aspects of biology or information theory, but his grasp of history of evolutionary theory, and of the crucial assumptions of the theory, appears to be poor. If the individual steps are not selectable, the classical Darwinian approach must be abandoned. This is what Behe and Denton have stressed. But Petrushka does not take the time to read Behe and Denton. He contents himself with rumor and hearsay about what they think. He also appears to have little knowledge of the actual writings of Darwin, Goldschmidt, Gaylord Simpson, Shapiro, etc. But of course we are all used to this here. Our Darwinian combatants, with the exception of Allen MacNeill, aren't very scholarly. They prefer to argue from blogs and Wikipedia and other snippets they've read about evolutionary theory on the internet. We, on the other hand, take time to read entire books on the history of evolutionary theory, and entire books on the most recent developments in evolutionary theory. This is why we know their position much better than they know ours, and why we will always prevail. One can't argue about what one hasn't studied. T.Timaeus
December 20, 2011
December
12
Dec
20
20
2011
01:02 PM
1
01
02
PM
PDT
To say that a species 'needs an adaptation' implies a target. To describe their extinction as a "failure to achieve" it emphasizes it. Just to clarify - are we to understand that certain spiders were on their way to extinction but avoided it by adapting (in countless unspecified genetic, phenotypic, and behavioral changes) to weave, monitor, inhabit, and feed from orbital webs? That just doesn't seem like a reasonable explanation when other creatures supposedly "adapted" just be getting bigger, faster, and stronger. Isn't that a much simpler change, since it doesn't require evolution to plan as many complex, coordinated changes which require foresight and purpose it didn't possess? That's absurd. And, when you maintain that it must have happened without any idea how and despite good reason to think it couldn't have, it's clear that you're leading with the conclusion rather than following the evidence to it. If a ground-dwelling spider's best chance at avoiding extinction was to develop the numerous changes to become an orbital web-weaving spider (even with several transitional states invented as required to fit the evidence to the conclusion) then why wouldn't it just become extinct? If, on the other hand, they could accomplish something so spectacular and complex to avoid extinction, how does anything ever go extinct? The features of living things are real. Extinction is real. The story of needing and achieving or not achieving adaptations is a "narrative gloss" with no substance.ScottAndrews2
December 20, 2011
December
12
Dec
20
20
2011
05:45 AM
5
05
45
AM
PDT
The most common result of a species needing an adaptation is failure to achieve the adaptation, and eventual extinction. The simpler the organism and the more prolific, the less likely to become extinct. the more specialized the organism, the more likely to go extinct. Bacteria and viruses are the dominant life forms because their population size and reproductive rate favors finding adaptations.Petrushka
December 19, 2011
December
12
Dec
19
19
2011
12:56 PM
12
12
56
PM
PDT
1 2

Leave a Reply