The Panda's Thumb Legacy Archive — pandasthumb.net

Signature in the Cell: self-contradiction and repetition

Posted 31 December 2009 by Nick Matzke

Of late the IDists have been complaining about the dearth of reviews by ID skeptics of Stephen Meyer's book Signature in the Cell. I agree, it would be nice if there were more reviews out there, but (a) the arguments boil down to the same old fallacious "improbability of assembly of functional sequence all at once from scratch by brute chance" creationist argument that dates back to at least the 1960s creation science literature, and (b) the book is tedious and repetitive, basically making the same unsupported assertions again and again in slightly different ways. I.e. information comes from intelligence and is too improbable to explain by chance, therefore intelligence! The actual known origin of the vast majority of genetic "information" -- DNA duplication followed by mutation and selection is (1) almost completely ignored by Meyer and (2) directly refutes Meyer's key claim, which is that the only known explanation of new information is intelligence. So in one sense, there is not a heck of a lot to review in Meyer's book. If you are a sufficient wonk about the ID debate, there is some interesting stuff about Meyer's highly revisionist account of his own history and the history of the ID movement, and there is an interesting study to be made of the science that Meyer left out of his book, but that makes for a big project, so it will be awhile before I or someone else get it out there. But, while reading across the book, you do occasionally come across some examples of truly bizarre argumentation. Here an example which I just posted in response to a Telic Thoughts challenge: There are many problems with the argument in Signature in the Cell, here is one. Meyer says that "information" -- sequence-specific function -- is densely concentrated in the DNA genome:
"Thus, far from being dispersed sparsely, haphazardly, and inefficiently within a sea of nonfunctional sequences (one that supposedly accumulated by mutation), functional genetic information is densely concentrated on the DNA molecule." (p. 461)
"Far from containing a preponderance of "junk" -- nonprotein-coding regions that supposedly perform no function -- the genome is dominated by sequences rich in functional information." (p. 461)
Furthermore, says Meyer, not only is this established truth, but it is a prediction of ID theory, and furthermore it was predicted by ID advocates a decade or more ago:
"The genome does display evidence of past viral insertions, deletions, transpositions, and the like, much as digital software copied again and again acumulates errors. Nevertheless, the vast majority of base sequences in the genome, and even the many sequences that do not code for proteins, serve essential biological functions. Genetic signal dwarfs noise, just as design advocates would expect and just as they predicted in the early 1990s." (p. 461)
However, at numerous places in the book, Meyer notes (correctly) that repetitive sequences have little information:
"Since information and improbability are inversely related, high-probability repeating sequences like ABCABCABCABCABCABC have very little information (either carrying capacity or content). And this makes sense too. Once you have seen the first triad of ABCs, the rest are "redundant"; they convey nothing new. They aren't informative. Such sequences aren't complex either. Why? A short algorithm or set of commands could easily generate a long sequence of repeating ABCs, making the sequence compressible." (p. 107)
Unfortunately for Meyer, he seems to not realize that 40-50% of the human genome (and most animal genomes of similar size) consists of LINEs, SINEs, segmental duplications, and other repeating elements. As documented here: http://www.ncbi.nlm.nih.gov/bookshelf/br.fcgi?book=hmg&part=A642
In other words, there is no way that in "the vast majority" of the genome genetic information is "densely concentrated" -- as proven by his own arguments! QED.

85 Comments

Nick (Matzke) · 31 December 2009

That chart may be a little out of date. Quoth wikipedia:
Retrotransposons (also called transposons via RNA intermediates) are genetic elements that can amplify themselves in a genome and are ubiquitous components of the DNA of many eukaryotic organisms. They are a subclass of transposon. They are particularly abundant in plants, where they are often a principal component of nuclear DNA. In maize, 49-78% of the genome is made up of retrotransposons[1]. In wheat, about 90% of the genome consists of repeated sequences and 68% of transposable elements[2]. In mammals, almost half the genome (45% to 48%) comprises transposons or remnants of transposons. Around 42% of the human genome is made up of retrotransposons while DNA transposons account for about 2-3%[3]. [...] 1. SanMiguel P, Bennetzen JL (1998). "Evidence that a recent increase in maize genome size was caused by the massive amplification of intergene retrotranposons". Annals of Botany 82 (Suppl A): 37–44. http://aob.oxfordjournals.org/cgi/reprint/82/suppl_1/37.pdf. 2. Li W, Zhang P, Fellers JP, Friebe B, Gill BS (November 2004). "Sequence composition, organization, and evolution of the core Triticeae genome". Plant J. 40 (4): 500–11. doi:10.1111/j.1365-313X.2004.02228.x. PMID 15500466. http://www.blackwell-synergy.com/doi/abs/10.1111/j.1365-313X.2004.02228.x. 3. Lander ES, Linton LM, Birren B, et al. (February 2001). "Initial sequencing and analysis of the human genome". Nature 409 (6822): 860–921. doi:10.1038/35057062. PMID 11237011. http://www.nature.com/nature/journal/v409/n6822/full/409860a0.html.

Glen Davidson · 31 December 2009

Thanks for that. So the book is much like the genome, filled with mistakes and repetitive junk. Trouble is, Meyer is intelligent enough to have made something better. Evolution has the excuse that it lacks rationality and intelligence, hence can only produce something like the genome. One thing that I think is too missing in these discussions is that the differences between the genome and software--which are in fact considerable--generally are what make the former evolvable. If he dared to consider that, he'd have to admit that life fits the predictions of evolution, not any entailed design criteria. Which is why he doesn't. Glen Davidson

http://tinyurl.com/mxaa3p

John Kwok · 31 December 2009

Nick and Glen,

What I found rather odd in Meyer's presentation is his false dichotomy between "experimental" and "historical" sciences, claiming that biology belongs to the latter. Of course he delves into extensive quote mining, citing none other than Stephen Jay Gould to make his case. Of course if this was really true, it wouldn't account for some of the landmark experiments in evolutionary biology (e. g. Lenski's and Endler's) recounted in recent books from Carl Zimmer and Richard Dawkins.

Joe Felsenstein · 31 December 2009

Nick wrote in the post: the book is tedious and repetitive, basically making the same unsupported assertions again and again in slightly different ways. I.e. information comes from intelligence and is too improbable to explain by chance, therefore intelligence!
(I should hasten to say that I have not more than skimmed the book). Is Meyer relying for this on William Dembski's Law of Conservation of Complex Specified Information? He should be aware that the LCCSI has been shown by Jeffrey Shallit (in Elsberry and Shallit, p. 27) to be unproven. I have also argued that the LCCSI cannot be proven to apply if we use the same specification (say asking whether the genotype has achieved some high level of fitness) before and after the evolutionary process acts. And to use the LCCSI to show that evolution is ineffective at putting information into the genome, you do have to keep the specification before and after the same). If Meyer is relying on Dembski's argument, he has built a castle on sand.

John Kwok · 31 December 2009

Joe I believe he is and he is ignoring all criticisms of Dembski's work, with the notable exception of an Elsberry and Wilkins paper in Biology and Philosophy, criticism from Pennock, Scott and Branch, claiming that Dembski's Design Inference does not depend on an "Argument from Ignorance". After just perusing the bibliography now, one not acquainted with the activities of the Discovery Institute might conclude that there are a few dissenters like Ken Miller, Robert Pennock, Genie Scott and Glenn Branch who enjoy taking potshots at Meyer, his colleagues and their work. This remarkable sense of denial of any meaningful criticism is almost as bad as Behe's in "The Edge of Evolution":
Joe Felsenstein said:
Nick wrote in the post: the book is tedious and repetitive, basically making the same unsupported assertions again and again in slightly different ways. I.e. information comes from intelligence and is too improbable to explain by chance, therefore intelligence!
(I should hasten to say that I have not more than skimmed the book). Is Meyer relying for this on William Dembski's Law of Conservation of Complex Specified Information? He should be aware that the LCCSI has been shown by Jeffrey Shallit (in Elsberry and Shallit, p. 27) to be unproven. I have also argued that the LCCSI cannot be proven to apply if we use the same specification (say asking whether the genotype has achieved some high level of fitness) before and after the evolutionary process acts. And to use the LCCSI to show that evolution is ineffective at putting information into the genome, you do have to keep the specification before and after the same). If Meyer is relying on Dembski's argument, he has built a castle on sand.

John Kwok · 31 December 2009

I might also add that Shallitt's critiques have been ignored by Meyer in this book (nor is Shallitt referred to at all, even in the book's extensive footnotes).

Mike Elzinga · 31 December 2009

Of late the IDists have been complaining about the dearth of reviews by ID skeptics of Stephen Meyer’s book Signature in the Cell. I agree, it would be nice if there were more reviews out there, but (a) the arguments boil down to the same old fallacious “improbability of assembly of functional sequence all at once from scratch by brute chance” creationist argument that dates back to at least the 1960s creation science literature, and (b) the book is tedious and repetitive, basically making the same unsupported assertions again and again in slightly different ways. …

— Nick Matzke
Looking back over the origins of the misconceptions and misrepresentations from creation “science” that have simply been taken over by ID, I have noticed that the central misconception that keeps popping up is the notion that all atomic and molecular structures come together from a uniform random sampling of atoms and molecules from essentially infinite sample spaces. Duane Gish was already bludgeoning biology teachers with this misconception back in the late 1960s and 1970s in Kalamazoo, Michigan. After the Institute for Creation “Research” was formed, as I recall, both Gish and Henry Morris were conflating entropy with disorder. Somewhere in that time period they started using the tornado-in-a- junkyard shtick. I was aware at that time of some of the misconceptions about entropy and the second law of thermodynamics floating around in physics, but chemist Frank L. Lambert appears to have been fighting the battle of these misconceptions in chemistry textbooks for quite some time now; perhaps even longer. If I understand his historical cataloguing of these misconceptions, I wonder if Gish got his own miseducation from such a textbook, and that reinforced some unfortunate statements about entropy by well-known physicists. At any rate, the misconceptions were under some control within the physics community as careful instructors made clear distinctions about what was being enumerated in counting available energy microstates. However, after the political campaign of the creationists got going in earnest, those erroneous memes spread like wildfire; and even scientists who debated creationists fell for them. Now these misconceptions are standard fare in all ID/creationist literature; and invariably they lead to the erroneous conclusions we see in all the work of Meyer, Dembski, Behe, Abel, and others. They are so predictable that I have often been able to “psych out” the path of their writings before I actually pick up one of their papers or books and slog through it. And they never correct themselves or retract anything, even after something like 40 years of attempted corrections by the scientific community. It’s getting harder and harder to believe ID/creationists like Meyer don’t know they are screwing up fundamental scientific concepts. But then, they are pretty wrapped up in themselves.

Nick (Matzke) · 31 December 2009

Yeah, Meyer explicitly relies on Dembski's specified complexity argument -- to the point that, when he tells the history, Meyer sort of says that he co-invented it back in the early 1990s.

However, Dembski really makes 2 different arguments based on 2 different definitions:

1. Definition: Specified complexity is defined as complex (aperiodic) sequence that codes for meaning or function (which defines what is "specified", according to them). Argument: such sequences are wildly improbable (so improbable as to be impossible, 10^-150) to have arisen through chance or other natural processes (which boil down to chance, according to ID advocates).

2. Definition: Specified complexity is *defined* as a sequence that has a probability of 10^-150 or less under all natural processes. Argument: The existence of these sequences proves evolution doesn't work. This argument is typically concealed in vast reams of mostly pointless calculation.

Traditionally, Dembski would put forward #1, which is basically the "popular", but when challenged with counterexamples of e.g. evidence that new genes had originated through natural processes, he or his defenders would switch to #2, and assert that, sure, maybe natural processes produced those genes, but all that means is that those don't qualify as examples of specified complexity, and thus aren't actually counterexamples. #2 is obviously just a pointless tautology.

(A similar game is often played with irreducible complexity -- sometimes explicitly, as when Dembski invokes IC to head off the possibility that gradual selection-based processes could explain SC. Of course, this sort of manuever makes his entire argument a pointless gloss on IC.)

Anyway, back to Meyer: Meyer pretty much relies on the #1 version of specified complexity, equates it to information, and doesn't get much into the details. I think the Universal Probability Bound is mentioned, but FWIW many of Meyer's citations are to e.g. Axe's papers which (allegedly; they actually don't) show improbabilities on the order of 10^-77 or so which is far from 10^-150.

Nick (Matzke) · 31 December 2009

And yeah Meyer ignores all of the detailed criticism of Dembski which has been published in many places...

TR Gregory · 31 December 2009

The whole line of argumentation that ID predicted function for non-coding DNA in the early 1990s while "Darwinists" stubbornly assumed non-function is completely false historically. The standard assumption based on strict Darwinian principles in the 1970s and at least the early 1980s was that if non-coding sequences did not have a function, natural selection would have gotten rid of them. The selfish DNA hypothesis arose in the early 1980s as a response to the widespread assumptions of function, but even it did not deny the likely cooption of some transposable elements, especially in regulation. Meanwhile, discussions of satellite DNA and other sequences were largely about trying to determine their function, even after the selfish DNA papers. By 1994, there were stories in major journals stating that non-coding DNA was "long thought to have been non-functional but is now turning out to have important roles". I have done my best to document the actual status of non-coding DNA in the literature since the 1970s here. The dismissal of "junk DNA" never happened in any significant way (and even if it did it could only have been between about 1983-1994), and the reason is that adaptationist assumptions led people to expect functions for any DNA that is so abundant.

(And yes, the fig is way out of date -- it was drawn 10 years ago).

Nick (Matzke) · 31 December 2009

TT replies: http://telicthoughts.com/the-signature-in-the-cell-challenge/

John Kwok · 31 December 2009

Am quoting the following text from my Amazon.com review of "Signature", which, I believe, demonstrates the poor logic used by Meyer in asserting that Intelligent Design does produce "testable" scientific hypotheses:

So should we accept Meyer's proposition that Intelligent Design is a valid scientific theory simply because it produces testable hypotheses? What hypotheses? For example, he asserts on Page 489, "Design hypotheses envisioning discrete intelligent action also predict a pattern of fossil evidence showing large discontinuous or `quantum' increases in biological form and information at intervals in the history of life. Advocates of this kind of design hypothesis would expect to see a pattern of sudden appearance of sudden appearance of major forms of life as well as morphological stasis." Moreover, he claims "...they would also predict a `top-down' pattern of appearance in which large-scale differences in form (`disparity' between many separate body plans) emerge suddenly and prior to the occurrence of lower-level (i.e., species and genus) differences in form. Neo-Darwinism and front-loaded hypotheses expect the opposite pattern, a `bottom-up' pattern in which small differences in form accumulate first (differentiating species and genera from each other) and then only much later building to the large-scale differences in form that differentiate higher taxonomic categories such as phyla and classes."

Granted, life would be a lot simpler for paleontologists and paleobiologists if they heeded Meyer's most generous advice. We wouldn't have to worry about long-term persistence of ecological communities replete with morphological stasis of their constituent taxa over considerable spans of geological time or those unfortunate "accidents" known as mass extinctions which have "reshuffled the deck" that is Earth's biodiversity not just once, but at least seven times over the past five hundred fifty-odd million years. After each of these "accidents" we do see eventual recovery of the Earth's biosphere via the "bottom-up" pattern that Meyer so clearly disdains. What we don't see however, is any indication of some Intelligent Designer(s) acting to ensure some kind of restoration of our planet's biodiversity. All the patterns seen in the fossil record are due to natural laws and processes acting on populations of organisms, not through the direct intervention of Intelligent Designer(s) like Mother Goose, Yahweh or the Klingons.

John Kwok · 31 December 2009

Meyer also claims that examples of "poor design" in nature merely illustrate that "perfect" designs can "degenerate". So, hypothetically, I am certain he would assert that the panda's thumb is the direct consequence of some kind degeneration of an initial "perfect" design.

Miranda · 31 December 2009

"I am certain he would assert that the panda’s thumb is the direct consequence of some kind (sic) degeneration of an initial “perfect” design."

John, he might assert that, but first he'd want you to listen to this podcast about the panda's thumb (not this website :-) ):
http://intelligentdesign.podomatic.com/entry/eg/2009-04-29T14_35_14-07_00

John Kwok · 31 December 2009

Probably after I watch the annual New Year's Day concert performed each year in Vienna by the Vienna Philharmonic Orchestra. I'd rather listen to superb playing of Strauss waltzes and similar scores than hearing Meyer lying through his teeth, which he seems to do all the time, whether it is in print, in a documentary video (e. g. the newly released "Darwin's Dilemma) or in an evolution vs. creationism "debate". Until then, I wish you a most happy new year:
Miranda said: "I am certain he would assert that the panda’s thumb is the direct consequence of some kind (sic) degeneration of an initial “perfect” design." John, he might assert that, but first he'd want you to listen to this podcast about the panda's thumb (not this website :-) ): http://intelligentdesign.podomatic.com/entry/eg/2009-04-29T14_35_14-07_00

Alex H · 31 December 2009

Glen Davidson said: So the book is much like the genome, filled with mistakes and repetitive junk.
No, unlike Signature In the Cell, genomes actually have a purpose.

Wayne Robinson · 1 January 2010

Even if most of the DNA in the human genome has a function, it doesn't explain why the human genome has 3 billion base pairs, but that the lowly Amoeba dubia has 670 billion base pairs (220 times as many). Listened to Nelson's podcast, Miranda, I take it you have a bad hangover from last night, and you are trying to spread your misery.

TomS · 1 January 2010

Because the subtitle of the book is "DNA and the evidence for intelligent design", I skimmed through the book looking for evidence for intelligent design - or just a description of intelligent design, something that one could imagine evidence for.

Joe Felsenstein · 1 January 2010

TR Gregory said: The whole line of argumentation that ID predicted function for non-coding DNA in the early 1990s while "Darwinists" stubbornly assumed non-function is completely false historically. The standard assumption based on strict Darwinian principles in the 1970s and at least the early 1980s was that if non-coding sequences did not have a function. ... The dismissal of "junk DNA" never happened in any significant way (and even if it did it could only have been between about 1983-1994), and the reason is that adaptationist assumptions led people to expect functions for any DNA that is so abundant.
Thanks, that site of yours has a great collection of quotes from the literature. Two major groups rejecting the notion of junk DNA were (1) panselectionists among evolutionary biologists (as you have noted), and (2) molecular biologists, many of whom have tended to see the genome as one big machine, with all that extra DNA just waiting to have its function discovered if only more and more grant money could be obtained. In fact population geneticists very early on were worried that there would be too much mutational load if all that extra DNA had function and thus its sequence had to be conserved. You know the literature -- am I summarizing incorrectly? But there may be even one more way on which ID advocates are fudging the historical record. Where are their scientific predictions that there is no junk DNA? They do frequently assert that there is none, but where in “ID theory” does that “prediction” come from? As far as I can see it comes from theological arguments about what an omniscient deity would do. I cannot see anything in the negative scientific arguments of Dembski and of Behe that speaks to the issue.

DS · 1 January 2010

Joe wrote:

"But there may be even one more way on which ID advocates are fudging the historical record. Where are their scientific predictions that there is no junk DNA? They do frequently assert that there is none, but where in “ID theory” does that “prediction” come from? As far as I can see it comes from theological arguments about what an omniscient deity would do. I cannot see anything in the negative scientific arguments of Dembski and of Behe that speaks to the issue."

Precisely. As soon as creationists start making statements about the identity, motives or abilities of the creator, they automatically open themselves up to falsification. This is an especially pernicious problem because they often have no idea of the evidence that is already available. Why they can't be bothered to do a little research is beyond me (almost).

If they claim that there is no "junk DNA", then the claim is falsified by present knowledge and the burden of proof is on them to demonstrate a function for every base pair. Without a laboratory, that could be difficult. Of course, even if they can demonstrate this, they would then have to demonstrate the function every vestigial organ as well, since their claim apparently rest on the premise that the creator would not create useless things for no reason.

And of course even this wouldn't really help the creationist cause, because they would still have to explain all of the other genetic, developmental and morphological patterns that are consistent with common descent and inconsistent with any creation scenario.

Man, no wonder they just throw up their hands, claim they have no idea who the creator is and refuse to make any testable predictions. Every time they do, they get slammed.

TR Gregory · 1 January 2010

Joe Felsenstein said: But there may be even one more way on which ID advocates are fudging the historical record. Where are their scientific predictions that there is no junk DNA? They do frequently assert that there is none, but where in “ID theory” does that “prediction” come from? As far as I can see it comes from theological arguments about what an omniscient deity would do. I cannot see anything in the negative scientific arguments of Dembski and of Behe that speaks to the issue.
Absolutely right. Maybe they made this "prediction" (read: the same assumption as adaptationists) in their own literature, but I haven't been able to get a clear explanation from current proponents of why they should expect this. Human design generally includes some imperfections, redundancies, superfluous components, etc., so why should design per se imply no junk DNA? It must be a theological claim, rarely stated explicitly.

OgreMkV · 1 January 2010

This paper: http://www.sciencemag.org/cgi/content/full/291/5507/1304

Has some very interesting comments, mainly in Table 11. Genome overview.

Percent of human genome that is repeats: 35
Most gene-rich chromosome Chr. 19 (23 genes/Mb)
Least gene-rich chromosomes Chr. 13 (5 genes/Mb), Chr. Y (5 genes/Mb)

Looks like that harms Meyer's case too.

TomS · 1 January 2010

TR Gregory said: Human design generally includes some imperfections, redundancies, superfluous components, etc., so why should design per se imply no junk DNA? It must be a theological claim, rarely stated explicitly.
Until we get a positive description of what "intelligent design" is, the most that we can go on is some analogy with the only design that we know about. The only design that we know anything about uses regularities (and sometimes chance, too). Try to design something without knowing what happens when you do something.

Opisthokont · 1 January 2010

Things are even worse for Meyer than the chart used in this post indicates: there are far less than "~80 000 genes" in the human genome! The number is something closer to 23 000 (although I do not know whether copy numbers of individual genes are taken into account in either figure). To be fair, the chart comes from a book published in 1999, when the number was assumed to be much higher than it turned out to be. Of course, either figure is contrary to Meyer's claim.

Miranda · 1 January 2010

John Kwok, the podcast is not of Meyer.

Glen Davidson · 1 January 2010

This is how I characterized their "prediction" regarding "junk DNA" on Coyne's blog:
But there’s little junk DNA in the genome, just like ID predicts. Cause it predicts, you know, whatever amount it turns out to be.
Or, as I predicted of Meyer's 12 "ID predictions" in an appendix of his book, these predictions would just be something somebody on the ID side said. How could they be entailed, given that they have no theory? Needless to say, the "predictions" are indeed just something somebody said, and most rely on the false dichotomy that if evolution (or the infant abiogenesis) doesn't explain it, Goddidit Designerdidit. What's ironic is not even so much that they claim to predict anything, but that they whine piteously whenever we bring up "poor design" (and no, the panda's thumb apparently isn't one of them) that we're doing theology. They didn't predict "good design" did they? (They didn't dare.) Except that they claim to have done just that with respect to "junk DNA," even though they'll never commit to any criteria, quantities, or characterization of "the designer" (we don't really say how much it should be, either, because we recognize that too much is unknown to predict, something they won't admit). Well, ID hypocrisy is hardly a newly discovered trait. BTW, why does spell check flag American spellings of words, when this is an American-based site? It likes "characterisation" and "realise", but not "characterization" and "realize." Glen D

http://tinyurl.com/mxaa3p

TomS · 1 January 2010

Glen Davidson said: BTW, why does spell check flag American spellings of words, when this is an American-based site? It likes "characterisation" and "realise", but not "characterization" and "realize."
Aha - a perfect opportunity for applying the methodology of ID: Why does spell check do such-and-such? Because it is intelligently designed. Does anyone need any more "explanation"?

John Kwok · 1 January 2010

Yeah, it's of his YEC colleague Paul Nelson trying to refute Gould's observation (which IMHO is still valid, if you wish to compare a panda's thumb with ours). Just heard it and am now looking forward to hearing some really good music courtesy of the Wiener Philharmoniker (Vienna Philharmonic Orchestra) and their New Year's Day Concert conductor, French conductor Georges Pretre:
Miranda said: John Kwok, the podcast is not of Meyer.

Nick (Matzke) · 1 January 2010

But there may be even one more way on which ID advocates are fudging the historical record. Where are their scientific predictions that there is no junk DNA? They do frequently assert that there is none, but where in “ID theory” does that “prediction” come from? As far as I can see it comes from theological arguments about what an omniscient deity would do. I cannot see anything in the negative scientific arguments of Dembski and of Behe that speaks to the issue.
Meyer does cite some ID stuff from the 1990s that he says shows they predicted that junk DNA wasn't junk -- I haven't looked those up to double-check. But yeah, the overall argument is bunkum on so many levels. (1) there is no ID theory in the first place, (2) even we interpret the statement "ID did something somewhere in the Universe" as a theory, that doesn't tell you anything about junk DNA, (3) what really predicts function for e.g. the noncoding DNA in the human genome is "God specially created humans relatively recently, and He wouldn't have put in a bunch of low-information junk sequence, because we know (somehow) that he's a neat guy" -- but ID "officially" denies that they are talking about God, that they make any such assumptions, or that they deny common ancestry, even though they clearly do. Only after all this do we get to my point, (4), repetitive DNA shows that Meyer can't be right about information density anyway, on his own arguments. But yeah, I wouldn't take #4 to be a falsification or test of "ID", "ID" as officially presented is untestable and not science. It is some kind of test of #3, though, which has some validity when made explicit. My biggest point, though, is that #4 is strong evidence for proposition #5: (5) The ID movement and its leading proponents, and fans, are clueless, can't think critically, can't detect or correct simple errors, and generally are completely intellectually bankrupt.

Nick (Matzke) · 1 January 2010

But there may be even one more way on which ID advocates are fudging the historical record. Where are their scientific predictions that there is no junk DNA? They do frequently assert that there is none, but where in “ID theory” does that “prediction” come from? As far as I can see it comes from theological arguments about what an omniscient deity would do. I cannot see anything in the negative scientific arguments of Dembski and of Behe that speaks to the issue.
Meyer does cite some ID stuff from the 1990s that he says shows they predicted that junk DNA wasn't junk -- I haven't looked those up to double-check. But yeah, the overall argument is bunkum on so many levels. (1) there is no ID theory in the first place, (2) even we interpret the statement "ID did something somewhere in the Universe" as a theory, that doesn't tell you anything about junk DNA, (3) what really predicts function for e.g. the noncoding DNA in the human genome is "God specially created humans relatively recently, and He wouldn't have put in a bunch of low-information junk sequence, because we know (somehow) that he's a neat guy" -- but ID "officially" denies that they are talking about God, that they make any such assumptions, or that they deny common ancestry, even though they clearly do. Only after all this do we get to my point, (4), repetitive DNA shows that Meyer can't be right about information density anyway, on his own arguments. But yeah, I wouldn't take #4 to be a falsification or test of "ID", "ID" as officially presented is untestable and not science. It is some kind of test of #3, though, which has some validity when made explicit. My biggest point, though, is that #4 is strong evidence for proposition #5: (5) The ID movement and its leading proponents, and fans, are clueless, can't think critically, can't detect or correct simple errors, and generally are completely intellectually bankrupt.

snaxalotl · 1 January 2010

Nick (Matzke) said: Anyway, back to Meyer: Meyer pretty much relies on the #1 version of specified complexity, equates it to information, and doesn't get much into the details. I think the Universal Probability Bound is mentioned, but FWIW many of Meyer's citations are to e.g. Axe's papers which (allegedly; they actually don't) show improbabilities on the order of 10^-77 or so which is far from 10^-150.
another way of describing what Meyer does, is that he pumps his audience's intuition by giving examples of intended function until they conflate function and intended function reflexively. Then he points to biological examples of function, and says "look: god".

John Kwok · 1 January 2010

That's a most cynical way of perceiving it, but of course, you are absolutely right:
snaxalotl said:
Nick (Matzke) said: Anyway, back to Meyer: Meyer pretty much relies on the #1 version of specified complexity, equates it to information, and doesn't get much into the details. I think the Universal Probability Bound is mentioned, but FWIW many of Meyer's citations are to e.g. Axe's papers which (allegedly; they actually don't) show improbabilities on the order of 10^-77 or so which is far from 10^-150.
another way of describing what Meyer does, is that he pumps his audience's intuition by giving examples of intended function until they conflate function and intended function reflexively. Then he points to biological examples of function, and says "look: god".

Olorin · 1 January 2010

Stephen Meyer demands to know where biological information comes from. Information, he claims, can only come from an intelligence, and the existence of biological information proves that this intelligence exists.

Permit me to make a similar claim. That nature is beautiful, and that beauty springs only from an intelligence. Tonight’s sunset, for example, contained 23.4 milliHelens of beauty.[1] Mindless, naturalistic weather patterns cannot create beauty. Every time we find beauty---in a painting, a symphony, whatever---we find an intelligence that creates it. Therefore, the existence of beauty in nature proves the existence of an intelligence for its origin.

==================
[1] A milliHelen is enough beauty to launch 1 ship.

DS · 1 January 2010

Olorin wrote:

"Stephen Meyer demands to know where biological information comes from. Information, he claims, can only come from an intelligence, and the existence of biological information proves that this intelligence exists."

Well he can demand whatever he wants, but if he wants to know where information comes from, then why does he claim that it can only come from intelligence? That's just plain crazy. A single example is sufficient to falsify that hypothesis. Intelligence is not required in order to produce information, intelligence is only required in order to interpret information. Why can't he understand this? What is so hard to understand?

Is there information in the color of your urine? What intelligence created that information?

Is there information in your white cell count? What intelligence created that information?

Is there information in the color of a maple leaf? What intelligence created that information?

Is there information in the width of tree rings? What intelligence created that information?

Is there information in the position of SINE insertions in cetacean genomes? What intelligence created that information?

John Kwok · 1 January 2010

The biological information you seek came from the Klingons. Just received a transmission from the current chancellor of the Klingon Empire and he told me so. They used Admiral Kirk's "slingshot" technique to travel back in time and seed the primordial Earth with microbial life over 4.1 Billon Years ago:
Olorin said: Stephen Meyer demands to know where biological information comes from. Information, he claims, can only come from an intelligence, and the existence of biological information proves that this intelligence exists. Permit me to make a similar claim. That nature is beautiful, and that beauty springs only from an intelligence. Tonight’s sunset, for example, contained 23.4 milliHelens of beauty.[1] Mindless, naturalistic weather patterns cannot create beauty. Every time we find beauty---in a painting, a symphony, whatever---we find an intelligence that creates it. Therefore, the existence of beauty in nature proves the existence of an intelligence for its origin. ================== [1] A milliHelen is enough beauty to launch 1 ship.

Arthur Hunt · 1 January 2010

A review that cuts the heart out of Meyer's book. A snippet from the abstract:

"Accordingly, there was likely a stereochemical era during evolution of the genetic code, relying on chemical interactions between amino acids and the tertiary structures of RNA binding sites."

So much for Meyer's idea that the genetic code is a designed, arbitrary mapping. So much for all 600+ pages of his book, based as it is on something that is contradicted by experimental evidence.

fnxtr · 1 January 2010

Clever, Olorin. Stealing it.

Mike Elzinga · 1 January 2010

Arthur Hunt said: A review that cuts the heart out of Meyer's book. A snippet from the abstract: "Accordingly, there was likely a stereochemical era during evolution of the genetic code, relying on chemical interactions between amino acids and the tertiary structures of RNA binding sites." So much for Meyer's idea that the genetic code is a designed, arbitrary mapping. So much for all 600+ pages of his book, based as it is on something that is contradicted by experimental evidence.
Even better, it shows just how silly all ID/creationist literature is about how atoms and molecules interact. All the hokum about “spontaneous molecular chaos”, “irreducible complexity”, “complex specified information”, “genetic entropy”, violations of thermodynamics, and the rest looks extremely stupid when compared with well-known physical and chemical phenomena such as

Such principles, which we summarize in a “polar profile”, are useful in explaining newly selected specific RNA binding sites for free amino acids bearing varied side chains charged, neutral polar, aliphatic, and aromatic.

Then there are hydrophobic and hydrophilic properties of chemical – especially organic – compounds; there are Van der Waals interactions, dimer and polymer, dipole-dipole and multipole interactions, the list goes on and on. When reading ID/creationist books, one gets the distinct impression that, despite their so-called “advanced degrees”, the authors of these books haven’t even mastered 8th grade general science. They don’t even understand what is going on when their eyeglasses fog up.

TomS · 2 January 2010

I suggest some cases of "complex specified information", if there is anything which answers this expression:

The close physical relationship which the human body bears to those of chimps and other apes.

Ecological relationships, such as between predator and prey, parasitism.

Changes to the world of life over time, including extinctions and speciations.

Of course, because ID is not a theory, they can shut off discussion whenever they want. But perhaps we can make it abundantly clear that this is what they are doing.

Sylvilagus · 2 January 2010

Arthur Hunt said: A review that cuts the heart out of Meyer's book. A snippet from the abstract: "Accordingly, there was likely a stereochemical era during evolution of the genetic code, relying on chemical interactions between amino acids and the tertiary structures of RNA binding sites." So much for Meyer's idea that the genetic code is a designed, arbitrary mapping. So much for all 600+ pages of his book, based as it is on something that is contradicted by experimental evidence.
Would someone please offer a translation and/or clarification of the significance of this finding for Meyer's argument? I am interested but the material is way outside my knowledge base. Thanks.

Frank J · 2 January 2010

Of course, because ID is not a theory, they can shut off discussion whenever they want. But perhaps we can make it abundantly clear that this is what they are doing.

— TomS
It's possible that I missed it, but what frustrates the heck out of me, and probably you too, is the dearth of questions of where (in vivo? new cells from scratch?), when and how those "designs" were inserted. I know that Meyer concedes that the Cambrian "explosion" was ~530 million years ago, but I also know that he doesn't like to advertise it to his YEC and young-life OEC fans. If we keep the subject on whether something is designed, or how evolution explains those "designs" they don't have to shut off the "discussion," because that hands them more data and quotes to misrepresent. But when we ask the simple questions they hate to answer, all but their most hopeless fans can see that they're trying to hide something.

Rich · 2 January 2010

Time for a computer engineer with experience with hard drive read channels to enter the fray.
“Since information and improbability are inversely related, high-probability repeating sequences like ABCABCABCABCABCABC have very little information (either carrying capacity or content). And this makes sense too. Once you have seen the first triad of ABCs, the rest are “redundant”; they convey nothing new. They aren’t informative. Such sequences aren’t complex either. Why? A short algorithm or set of commands could easily generate a long sequence of repeating ABCs, making the sequence compressible.” (p. 107)
As Nick has already noted Meyer is correct and is just restating the inferences from the work of Kolmogorov and Chaitin, namely a repetitive string is easily compressible in a lossless manner and thus has low Kolmogorov complexity. The bolded part is important as we shall see in a second. The first sentence is just wrong in that a repetitive sequence doesn't show low improbability since improbability is supposed to be the hallmark of intelligent design. Consider redundant arrays of inexpensive discs (RAID). Here the designer deliberately puts the same information on different drives in order to be fault tolerant. Seeing repetitive information in the string may show design rather than the lack of it. In fact, one could say that the repetitive elements show a kind of fault tolerance in the genome. A deadly missense mutation becomes a neutral nonsense violation which can accumulate via genetic drift. Later, these mutations then can become available for positive selection. Thus, even true "junk" DNA serve a purpose in making life robust against harmful mutations and in the end make beneficial mutations more likely. The problem of this kind of lower-case ID argument is that a) it accepts evolution as fact and ID is more concerned with disproving evolution than proving intelligent design and b) because it accepts evolution as fact the alternative that the genome is not designed is just as valid a hypothesis. In other words, you cannot make a scientific inference of design because a differential test between the two hypotheses cannot be made. (This is the kind of prediction that needs to be made in the appendix but sadly is not.) Still, this is not the largest error made from Nick's quotes of Meyer.
The genome does display evidence of past viral insertions, deletions, transpositions, and the like, much as digital software copied again and again acumulates errors. [Emphasis mine]
Remember my bolded word lossless? This is because unlike compressing photographs compressing program text must not have any changes. While RAID systems are used for a high-reliability disk farms for most uses is uncompetitive. The reason why digital has won over analog is that you can copy the information over and over again and not lose anything. What the drive manufacturers want is smaller and more dense storage. This means they cannot afford redundant information. Thus, special codes are used to add as little redundant information as possible. Furthermore, iterative decoding is done where the redundancy is added in time in addition to space. That's the point of a read channel, to extract the signal from multiple discs and multiple reads from the noise which is analogous to random mutation. What this means is a digital system does not accumulate errors because the errors are removed in each read/write cycle. But life does not have a read channel. It just makes copies and accumulates errors just a Meyer noted. From this we can see that not only is Dembski's information theory busted but the entire analogy used by ID is also busted. When life reproduces it does it only once so no iterative decoding there. The genome also accumulates errors and uses these errors to create new functions. A computer program stored on a disc is invariant by design and does not evolve. If the program does change then it is a design flaw. As I stated above one could make the argument that evolution is another way of doing design but that's not what ID argues. By being so violently anti-evolution ID potentially misses the real signature in the cell for a forged one.

Mike Elzinga · 2 January 2010

Rich said: From this we can see that not only is Dembski's information theory busted but the entire analogy used by ID is also busted. When life reproduces it does it only once so no iterative decoding there. The genome also accumulates errors and uses these errors to create new functions. A computer program stored on a disc is invariant by design and does not evolve. If the program does change then it is a design flaw. As I stated above one could make the argument that evolution is another way of doing design but that's not what ID argues. By being so violently anti-evolution ID potentially misses the real signature in the cell for a forged one.
Nice analogy. This relates to the important point that evolution is not aiming at any specific target; what falls out is what it is. There isn’t anything to preserve or correct. There is yet another aspect to evolution of complex systems that precludes design; it happens everywhere at every level in nature. From protons and neutrons, to atoms molecules, liquids, solids, through all of organic chemistry up to living organisms, evolution and emergence have the same kinds of manifestations. And what becomes at each step, while dependent on lower-lying rules, can be extremely sensitive to contingencies yet robust enough to persist. Just because some humans with sectarian perspectives think that certain dendrites on the tree of evolution are the targets of evolution doesn’t make it so. Targeted evolution is not generally true in nature.

DS · 2 January 2010

Rich wrote:

"A computer program stored on a disc is invariant by design and does not evolve. If the program does change then it is a design flaw. As I stated above one could make the argument that evolution is another way of doing design but that’s not what ID argues. By being so violently anti-evolution ID potentially misses the real signature in the cell for a forged one."

Exactly. If biological systems were in fact designed, (they were not and certainly not intelligently), then they were designed to evolve. At the very best, that makes the entire evolution thing god's idea. Why try to deny this? Why try to place artificial limits on the ability of evolution to generate new features. Are they trying to say that god designed systems to to be suboptimal and to evolve poorly as well? What an incompetent boob.

Wheels · 2 January 2010

I like the RAID analogy; it shows how real design is different than "Intelligent Design." Think I'll add that explanation to my list of how ID is useless compared to sciences that use design inferences (SETI and archeology).
Mike Elzinga said: When reading ID/creationist books, one gets the distinct impression that, despite their so-called “advanced degrees”, the authors of these books haven’t even mastered 8th grade general science. They don’t even understand what is going on when their eyeglasses fog up.
I've only got some entry-level college education about physics and chemistry, but some of the flaws in all the "random chance" arguments against abiogenesis were readily apparent to me and I could explain why they're wrong. For example, carbon has an affinity to bond with lots of other elements while the noble gases don't, metals and non-metals bond more readily than most any two metals, etc., and some types of bonds are favored above others, so there's no way a totally "random chance" model of chemistry (especially organic chemistry) could ever approximate the formation of amino acids, DNA, or cells. Since the different chemical tendencies of elements are generally taught in high school by the latest, I don't see how anybody with a degree in science or mathematics can possibly make the "random chance" argument for anything to do with the chemistry of life, but cdesign proponentsists make these arguments all the time.

Henry J · 2 January 2010

One way of looking at "intelligent design" is to note that an evolving gene pool does have some of the attributes associated with intelligence. It tries different things, and it "remembers" the things that worked, and emphasizes those that worked well. It lacks things like foresight, keeping track of mistakes to avoid in the future, and ways of redoing features that have undesirable side effects.

Henry

Mike Elzinga · 2 January 2010

Wheels said: I've only got some entry-level college education about physics and chemistry, but some of the flaws in all the "random chance" arguments against abiogenesis were readily apparent to me and I could explain why they're wrong.
Indeed you can; and you have often done so here on PT.

Since the different chemical tendencies of elements are generally taught in high school by the latest, I don’t see how anybody with a degree in science or mathematics can possibly make the “random chance” argument for anything to do with the chemistry of life, but cdesign proponentsists make these arguments all the time.

I still remember back in the late 1960s when I thought they were making honest mistakes that could be easily explained and corrected. It was a stinging slap-in-face when I observe them turning around a reusing the same misconceptions repeatedly in new venues after they had been shown they were wrong. That was when I realized they were just stinking dishonest. They haven’t changed. Even Fox Noise has picked up the same tactic; which tells you what cesspool they drink from.

Henry J · 2 January 2010

Since the different chemical tendencies of elements are generally taught in high school by the latest,

Yeah, chemistry is an elementary subject.

Olorin · 2 January 2010

Henry J said: One way of looking at "intelligent design" is to note that an evolving gene pool does have some of the attributes associated with intelligence.
It is surprising that no one has yet mentioned the theory of complex systems. When any system has components with certain levels of variety, connectedness, interdependence it can exhibit emerging properties such as evolution, novelty, adaptation---and even what we call “intelligence.” Show a creationist a few systems that exhibit novel behavior, yet demonstrably have no central controller, no overall guiding principle separate from the system itself. Could his dogma survive that? At least, without peals of laughter from his audience.

Mike Elzinga · 2 January 2010

Olorin said: Show a creationist a few systems that exhibit novel behavior, yet demonstrably have no central controller, no overall guiding principle separate from the system itself. Could his dogma survive that? At least, without peals of laughter from his audience.
As we have mentioned a number of times before, the fields of condensed matter physics and organic chemistry provide literally thousands of examples. Whenever any ID/creationist brings up “spontaneous molecular chaos”, or “irreducible complexity”, or “complex specified information”, or “genetic entropy”, or “entropy barriers” or anything else from their arsenal of pseudo-scientific “impossibilities”, one can simply point to the existence of liquids and solids as an obvious counter example to this crap. But, as I mentioned on another thread, they don’t even understand what is going on when their eyeglasses fog up. They’re as hydrophobic as a duck’s back.

Gary Hurd · 2 January 2010

Wheels said: I like the RAID analogy; it shows how real design is different than "Intelligent Design." Think I'll add that explanation to my list of how ID is useless compared to sciences that use design inferences (SETI and archeology).
You might also like reading my chapter, "The Explanitary Filter, Archaeology and Forensics," in "Why Intelligent Design Fails: A Scientific Critique of the New Creationism" edited by Matt Young, and Taner Edis (2004 Rutgers University Press). I woke-up this morning determined to write about Meyer's asinine book. I have a set of notes on pages 52, 206-208, 224-226, 334-335, and the associated footnotes and references. It is all related material about the early Archean environment, and the spontaneous formation of complex molecules. None of Meyer's citations are to current research, nearly all to very dated and refuted publications 20, or more years old. He even cribbed one of his 1998 papers word-for-word. I looked at the Amazon.com page for "Cigar in Meyer's Ass" and read some of the 5 star "reviews." I was so revolted that I did yard-work all day. We are out numbered by liars, and fools.

Gary Hurd · 2 January 2010

Arthur Hunt said: A review that cuts the heart out of Meyer's book. A snippet from the abstract: "Accordingly, there was likely a stereochemical era during evolution of the genetic code, relying on chemical interactions between amino acids and the tertiary structures of RNA binding sites." So much for Meyer's idea that the genetic code is a designed, arbitrary mapping. So much for all 600+ pages of his book, based as it is on something that is contradicted by experimental evidence.
Freaking great article. Thanks for the link Arthur.

Matt Ackerman · 2 January 2010

Rich said: While RAID systems are used for a high-reliability disk farms for most uses is uncompetitive. The reason why digital has won over analog is that you can copy the information over and over again and not lose anything.
and
What this means is a digital system does not accumulate errors because the errors are removed in each read/write cycle.
Actually: no. While you are correct that digital information can make error free copies, the fact that information is digital does not guarantee that error free copies are made. Meyers really is right when he states that digital information copied again and again has some rate of error accumulation. With an analog system the probability of making a perfect copy is exactly zero. With any digital system the probability is strictly less than one. While my knowledge of computers is somewhat limited, the loss of information from computer storage should happen with to distinct phases. The magnetic domains on a hard drive presumably lose their orientation at some intrinsic rate, likely related to the temperature the hard drive is operating at. Additionally, bits in ram are also subject to some inevitable, if very low, rate of spontaneous bit fliping. So, when you are copying data from one hard drive to another, you are really making two copies, a copy from the hard drive into ram, and a copy from ram onto a new hard drive. If the rate at which errors accumulate in ram is higher than the rate at which errors accumulate on hard drives, then the act of copying does introduce errors. Now, assuming a well programed computer, as you have specified with your example, then this error rate can be made arbitrarily low, since you can perform error checking to detect the rare events where bits are spontaneously fliped on the hard drive or in ram with some very high probability. However, since you are always, in the end, comparing a collection of items each of which has some probability of being in error, there will always be some arbitrarily low but non-zero probability that the act of copying data introduces errors into that data. Now, if that probability is lower than the rate at which errors enter the hard drive, then copying data will actually lower the rate at which errors accumulate, but you get the picture.
But life does not have a read channel. It just makes copies and accumulates errors just a Meyer noted.
Well, I don't really know what a read channel is, so I can't say that life has something exactly analogous to a read channel, however, our cells certainly do maintain redundant copies of genetic information and use those redundant copies to correct errors that are detected in any one given copy of the information. Each living thing has minimally two copies of each gene, since DNA is kept in a double helix. Most living things have four or more copies, and these copies really are used to correct errors in other copies of genes. I don't see how this is a drastically different situation than in RAID arrays, if I am understanding your analogy correctly. Granted, if certain specific errors are introduced in two copies of information simultaneously the cell might not be able to detect the error, but the same should be true of the RAID array as well. Anyway, I can pour beer all over my cells and they will continue copying information just fine. I'd like to see your fancy dancy RAID arrays pull that one off. ;)

Matt Ackerman · 2 January 2010

Sorry if I sounded at all confrontational Rich, I really liked your comment.

I'm a biologist and someone has to stick up for the absolutely mindboggling awesomeness of the nano-machines that compose us all.

Flint · 2 January 2010

I know a bit about this computer stuff, and Matt is essentially correct. What he left out (he may assume we all know it) is that it is much easier to detect errors than to correct errors. Disk storage technology has a block of data associated with each sector containing a lot of information about what's in that sector. From this, it's vanishingly unlikely that a copying (or reading) error will go undetected. Not zero, of course, but very small. Problem is, recovering the original requires full redundency of both the data and the metadata describing it (to know which copy is wrong). In cases where the metadata says BOTH are wrong, it's still possible to compare the two, and where differences are found, attempt all permutations of reconstructing one from the other and recreating the resulting metadata until the metadata match. And even still, it's theoretically possible for error to creep in, because certain pairs of bit-swaps produce identical metadata, though the underlying data is clearly now wrong.

The RAM involved in making copies between disks is generally not a problem for a couple of reasons. Modern RAM has one or two error identifying and correcting bits (and repeated errors in the same RAM cells cause modern systems to disable banks containing such cells), and multiple disk-compares where mismatches are found can statistically eliminate RAM as the culprit. Again not a zero probability of error, just very very small.

BUT except in the most extraordinary circumstances, none of these attempts to restore original data after errors are detected, are actually done. The philosophy is, such errors are so extremely infrequent, and the "retry" option so effective on the rare occasions when errors are detected, that we can just live with the current low error rate. The rate of physical media errors (on disks or in RAM) is MUCH smaller than the error rate introduced by logic errors in programs (which can be extremely subtle, involving multiple programs simultaneously interacting with incompletely understood hardware, and dependent on the data themselves, etc.)

Rolf Aalberg · 3 January 2010

DS said: Olorin wrote: "Stephen Meyer demands to know where biological information comes from. Information, he claims, can only come from an intelligence, and the existence of biological information proves that this intelligence exists." Well he can demand whatever he wants, but if he wants to know where information comes from, then why does he claim that it can only come from intelligence? That's just plain crazy. A single example is sufficient to falsify that hypothesis. Intelligence is not required in order to produce information, intelligence is only required in order to interpret information. Why can't he understand this? What is so hard to understand? Is there information in the color of your urine? What intelligence created that information? Is there information in your white cell count? What intelligence created that information? Is there information in the color of a maple leaf? What intelligence created that information? Is there information in the width of tree rings? What intelligence created that information? Is there information in the position of SINE insertions in cetacean genomes? What intelligence created that information?
Information? It is everywhere. Long ago, we had a small potted ivy plant in our living room. I was quite active as a radio amateur (Thanks to the ARRL Radio Amateurs Handbook for teaching me English) back then, and interpreted a sequence of leaves categorizing them as either 'big' or 'small' - translated to Morse code revealing our ivy crying H-E-L-P.

Rich · 3 January 2010

Matt Ackerman said:
Rich said: While RAID systems are used for a high-reliability disk farms for most uses is uncompetitive. The reason why digital has won over analog is that you can copy the information over and over again and not lose anything.
and
What this means is a digital system does not accumulate errors because the errors are removed in each read/write cycle.
Actually: no. While you are correct that digital information can make error free copies, the fact that information is digital does not guarantee that error free copies are made. Meyers really is right when he states that digital information copied again and again has some rate of error accumulation. With an analog system the probability of making a perfect copy is exactly zero. With any digital system the probability is strictly less than one. While my knowledge of computers is somewhat limited ...
Nick, it's interesting how ID proponents speak on subjects they don't have a clue. Whether it's speaking about biology when they don't know biology or climatology when they don't know climatology. Now it's telling me who has been designing computers and hard drives for over a quarter century how they are designed! Both of the responses are woefully ignorant as I will show shortly.
But life does not have a read channel. It just makes copies and accumulates errors just a Meyer noted.
Well, I don't really know what a read channel is, so I can't say that life has something exactly analogous to a read channel, however, our cells certainly do maintain redundant copies of genetic information and use those redundant copies to correct errors that are detected in any one given copy of the information. Each living thing has minimally two copies of each gene, since DNA is kept in a double helix. Most living things have four or more copies, and these copies really are used to correct errors in other copies of genes. I don't see how this is a drastically different situation than in RAID arrays, if I am understanding your analogy correctly. Granted, if certain specific errors are introduced in two copies of information simultaneously the cell might not be able to detect the error, but the same should be true of the RAID array as well. Anyway, I can pour beer all over my cells and they will continue copying information just fine. I'd like to see your fancy dancy RAID arrays pull that one off. ;)
Please note that this oversimplifies on how a RAID system works, e.g. there are error correcting codes on each of the drive's data and the system will re-read the data if the error is detected but not corrected in case the error was caused by noise in the reading process. Let's say there are three copies of a gene and three copies of a computer program on a hard drive. Let's say there is an error in one of the copies. The RAID system copies the file and the biological entity reproduces. The read channel of the hard drive controller and RAID firmware read the information off the three drives and find that one of them doesn't match the other two. It then takes the consensus from the two correct copies and: 1. Creates three copies of the correct code 2. Fixes the bad original copy In all cases the checksum is re-computed and verified that the copy is correctly made. This is repeated until a verified copy is made. In a real-world situation for five 9s reliability regular off site backups are made of all the data. In the biological example the bad copy of the gene is passed on to all the offspring. It doesn't have a read channel or RAID firmware that corrects the bad copies. That's why parents are checked for diseases such as Huntington's because the errors do get passed on. One of the ways macroevolution is proven is through analyzing common "bad" copies between closely related species. When a computer data file error is detected we don't check the previous drive because the error is presumably created de novo. You simply could not do any kind of cladistic analysis like you can for living things showing the history of the file copies.

Dave Lovell · 3 January 2010

Rich said: Nick, it's interesting how ID proponents speak on subjects they don't have a clue. Whether it's speaking about biology when they don't know biology or climatology when they don't know climatology. Now it's telling me who has been designing computers and hard drives for over a quarter century how they are designed! Both of the responses are woefully ignorant as I will show shortly. ...
Sorry Rich, but whilst everything in your description may well be correct, it does not contradict Matt's assertion. Error free copying is impossible. Your use of the term "five 9s reliability" is in itself an acceptance of this, event though you are designing for reliability many orders of magnitude greater. No matter how big you make the checksum, there is always a non-zero probability that two or more errors will result in an unchanged checksum result. And other failures can occur too. For example, what happens if your read head mechanically damages a tiny part of the drive surface immediately after a successful (i.e. data verifying) read of the data that used to be there?

Dave Wisker · 3 January 2010

Sylvilagus said:
Arthur Hunt said: A review that cuts the heart out of Meyer's book. A snippet from the abstract: "Accordingly, there was likely a stereochemical era during evolution of the genetic code, relying on chemical interactions between amino acids and the tertiary structures of RNA binding sites." So much for Meyer's idea that the genetic code is a designed, arbitrary mapping. So much for all 600+ pages of his book, based as it is on something that is contradicted by experimental evidence.
Would someone please offer a translation and/or clarification of the significance of this finding for Meyer's argument? I am interested but the material is way outside my knowledge base. Thanks.
The article discusses work which pretty conclusively shows that at least eight amino acids have a strong stereochemical affinity for tRNA molecules whose binding sites contain the codons or anticodons assigned to them in the genetic code. In other words, for these amino acids, the association between codon and amino acid, which Meyer and all Iders steadfastly insist cannot have originated naturally, has been shown to be the result of simple chemical affinity, not the result of assignmnent by an intelligent agent.

Bilbo · 3 January 2010

Even if all 20 amino acids are attracted to their appropriate codons, it doesn't show that intelligent design hasn't been at work. Afterall, the way it currently happens is that an aminoacyl-tRNA-synthetase first binds the amino acid, then binds it to the opposite end of the tRNA, furthest from the codon. Not exactly the way we would expect it to work. Why wouldn't the amino acids just bind to their codons on the mRNA? Why the need for two intermediate components -- the synthetase and the tRNA? Is it possible that a designer, noticing the attraction between the amino acids and their codons, and knowing that would increase the likelihood of the amino acids being the right neighborhood, designed the synthetases and tRNAs accordingly?

Glen Davidson · 3 January 2010

Yes, Bilbo, we know you can always save your "design" by saying "that's the way he did it."

As it makes actual sense only via evolution, why is it not better to say that "God made a universe in which such an arrangement would evolve?" Or, if you're not concerned about God, just say that such an arrangement is explicable in this universe due to evolution, without reference to God.

Glen D
http://tinyurl.com/mxaa3p

DS · 3 January 2010

Bilbo said: Even if all 20 amino acids are attracted to their appropriate codons, it doesn't show that intelligent design hasn't been at work. Afterall, the way it currently happens is that an aminoacyl-tRNA-synthetase first binds the amino acid, then binds it to the opposite end of the tRNA, furthest from the codon. Not exactly the way we would expect it to work. Why wouldn't the amino acids just bind to their codons on the mRNA? Why the need for two intermediate components -- the synthetase and the tRNA? Is it possible that a designer, noticing the attraction between the amino acids and their codons, and knowing that would increase the likelihood of the amino acids being the right neighborhood, designed the synthetases and tRNAs accordingly?
The amino acids cannot bind to the codons because then the codons could not bind to the anticodons. The point is that if the system were intelligently designed from scratch, there is no reason whatsoever why there should be any correspondence between the tRNA structure and the codons. If however, the system evolved from a much simpler beginning, then such nonrandom associations would be expected. The presence of aminoacyl-tRNA-synthetases means that the affinity may no longer be required. This is strong evidence that the system has evolved rather than being designed. Of course the designer could have done it this way for some unknowable reason or possibly to fool us. But the fact that these observations are completely congruent with all other evidence once again points directly to evolution rather than design.

Arthur Hunt · 3 January 2010

Gary Hurd said:
Arthur Hunt said: A review that cuts the heart out of Meyer's book. A snippet from the abstract: "Accordingly, there was likely a stereochemical era during evolution of the genetic code, relying on chemical interactions between amino acids and the tertiary structures of RNA binding sites." So much for Meyer's idea that the genetic code is a designed, arbitrary mapping. So much for all 600+ pages of his book, based as it is on something that is contradicted by experimental evidence.
Freaking great article. Thanks for the link Arthur.
And thanks for the kind remark, Gary. I decided to expand on the subject - just goes to show what a little bit of encouragement can lead to.

Dave Wisker · 3 January 2010

Bilbo said: Even if all 20 amino acids are attracted to their appropriate codons, it doesn't show that intelligent design hasn't been at work. Afterall, the way it currently happens is that an aminoacyl-tRNA-synthetase first binds the amino acid, then binds it to the opposite end of the tRNA, furthest from the codon. Not exactly the way we would expect it to work.
Bilbo, Yarus and his colleagues are careful to point out that the current translation system we see today is almost certainly NOT what it was like at the time the genetic code was developing.
Why wouldn't the amino acids just bind to their codons on the mRNA? Why the need for two intermediate components -- the synthetase and the tRNA?
That would be a better question for ID proponents to have to explain, if you ask me. What if the designer had a Rube Goldberg kind of approach to design? Then any weird setup would be acceptable.
Is it possible that a designer, noticing the attraction between the amino acids and their codons, and knowing that would increase the likelihood of the amino acids being the right neighborhood, designed the synthetases and tRNAs accordingly?
Anything is possible, with a designer whose nature, capabilities, and intent are completely unknown.

saxalotl · 4 January 2010

Dave Lovell said: ...Error free copying is impossible...
probably true, but a relevant point is that in the digital domain you can reduce the error rate below any arbitrarily specified amount with appropriate design. in the real world an error rate that is insignificant compared to other considerations (eg a higher probability of a chaos demon appearing as a coincidence of quantum fluctuations and killing everyone in the office) might reasonably be described as zero. this would be analogous to the difference in certainty between a logical fact and a scientific fact

Rich · 4 January 2010

saxalotl said:
Dave Lovell said: ...Error free copying is impossible...
probably true, but a relevant point is that in the digital domain you can reduce the error rate below any arbitrarily specified amount with appropriate design. in the real world an error rate that is insignificant compared to other considerations (eg a higher probability of a chaos demon appearing as a coincidence of quantum fluctuations and killing everyone in the office) might reasonably be described as zero. this would be analogous to the difference in certainty between a logical fact and a scientific fact
Also note that you can look at the encoding and from it you can infer that error-free copying is a design goal. If you look at how the genome copies you can infer the exact opposite. When a RAID system has multiple copies the number of copies is fixed. The genome on the other hand has copy number variation. (If the ID folks were really thinking they would have not insisted that the genome is mostly functional because the less dense the genome is the more robust it is to copying errors.) The fact you have multiple copies and such things as heat shock proteins and other chaperones that allow life to tolerate a bad copy but nothing is done to minimize the bad copies until there is a copy that is so bad that it causes the life to not be able to reproduce as well as its peers. Error-free copying is assumed in anti-virus software. If it sees the operating system change it assumes that the change was not intended by the designer. The software then looks for the "signature" of the virus and then reverts the operating system back to its original state. What happens to life when the analogous thing happens? In humans, endogenous retroviruses occupy about 1% of the genome, in total constituting ~30,000 different retroviruses embedded in each person's genomic DNA (Sverdlov 2000). There are at least seven different known instances of common retrogene insertions between chimps and humans. So, life copies these pieces presumably not intended by the designer along common descent. Maybe if life is designed to tolerate these virus infections that they are just inactivated, but no. It seems that a ERV helped in the function of the placenta in mammals. Here the env gene of the retrovirus was co-opted to allow the embryo to implant in the placenta without it being rejected as "foreign". This brings me back to my original point. We have "non-intended" code in our genomes that nevertheless produces an important function. If this is designed you need a designer that not only controls the encoding of the genome but also controls contingent viral infections. The only such designer must be God with a capital G. If life is designed it's designed in such a way that to make an analogy with the kind of design I do is to make a category error like Meyer did in his book. But this causes a problem to ID because such an ID theory would be blatantly unconstitutional to teach in the public schools and it uncovers that ID is not "just science". I happen to believe that there is such a God but I do not desire to impose my religious beliefs in high school science classes other than maybe let the students know that people such as myself exist and that people who believe that evolution is completely purposeless also exist. It is the wonder of the scientific method that we can work side-by-side to advance scientific knowledge. Note that this is precisely what the NAS did when publishing on Evolution and Creationism.

JohnK · 4 January 2010

Bilbo said: Even if all 20 amino acids are attracted to their appropriate codons, it doesn't show that intelligent design hasn't been at work. ... Why wouldn't the amino acids just bind to their codons on the mRNA?
The paper does not show that, say, isoleucine simply preferentially "binds to" a isolated AUU codon. In fact, a staple of creationist literature is that it was shown long ago there are no preferences at that simplistic level, thus allegedly no 'code evolution' possible. Instead this paper (and its refs) shows that, for example, random RNA sequences at least 22 nt long, with a UAUU motif somewhere internally, will coil up and collectively form a pocket that often preferentially binds isoleucine. In Fig. 7a, the presence of the entire surrounding 'DRT' is required for binding, not just the codon/anticodon. Instead this constant association eventually, thru a succession of events outlined in Part IV, results in the trailing AUU becoming a Ile codon, and the leading UAU an anticodon to the current AUA Ile codon. This is not the simplistic 1:1 AA-to-codons-on-a-string m-RNA in ribosome of Fig. 7c. (In fact Ile had all 4 AUx codons assigned until AUG was captured by Met much later.)
Is it possible that a designer, noticing the attraction between the amino acids and their codons, and knowing that would increase the likelihood of the amino acids being the right neighborhood, designed the synthetases and tRNAs accordingly?
The designer could make any synthetases and thus any code desired. Why would It care about affinities only exemplified in longer chained random RNA libraries?

DS · 4 January 2010

John wrote:

"The designer could make any synthetases and thus any code desired. Why would It care about affinities only exemplified in longer chained random RNA libraries?"

This is exactly the point. The affinities do not falsify a designer hypothesis. The designer could presumably do just about anything. That hypothesis is not falsifiable. However, the affinities are exactly what one would expect if the current system actually evolved from a simpler preexisting system. Coincidence? A wise man once said:

"I believe in coincidences, I just don't trust them"

And, once again, I was right.

KP · 4 January 2010

I was just thinking about the information content in genomes and wondering, can't the simple process of recombination increase information as well? Any individual undergoing meiosis can potentially produce many different haplotypes -- presumably different from the haplotypes that came from either of that individual's parents. Wouldn't this count as an increase in "information?" Has Meyer or Dembski addressed this, or is it too simplistic a criticism? Maybe someone here at PT has already dealt with this?

KP · 4 January 2010

ps. I know that is on a different scale than the RNA-codon discussion, but isn't it still an increase in information without the requirement of intelligence?

Frank J · 5 January 2010

The designer could presumably do just about anything. That hypothesis is not falsifiable.

— DS
So why do critics of ID/creationism spend 99+% of their time arguing "no designer needed for that," which only gives the pseudoscientist opponent more "gaps" to play with and/or more quotes to mine? Why" is it so rare to hear one of us reply with "OK, so what exactly do you think the designer(s) did and when did he/she/it/they do it? A skilled Gish-galloper can find an almost unlimited supply of cool sound bites to promote unreasonable doubt of evolution, but how many ways can they say "it's not our job to connect dots" without eventually making it clear to all but the most hopelessly deluded audiences (who don't need anti-evolution activists anyway) that they are playing games, and know that they don't have a prayer against evolution.

TomS · 6 January 2010

Herbert Spencer "The Development Hypothesis" Originally published in The Leader, March 20, 1852. Reprinted, slightly modified, in: Essays: Scientific, Political, and Speculative. Library Edition, containing Seven Essays not before republished, and various other Additions (London: Williams and Norgate, 1891). Vol. 1 Online at: http://oll.libertyfund.org/?option=com_staticxt&staticfile=show.php%3Ftitle=335&chapter=12313&layout=html&Itemid=27
If they have formed a definite conception of the process, let them tell us how a new species is constructed, and how it makes its appearance. Is it thrown down from the clouds? or must we hold to the notion that it struggles up out of the ground? Do its limbs and viscera rush together from all the points of the compass? or must we receive the old Hebrew idea, that God takes clay and moulds a new creature? If they say that a new creature is produced in none of these modes, which are too absurd to be believed, then they are required to describe the mode in which a new creature may be produced—a mode which does not seem absurd; and such a mode they will find that they neither have conceived nor can conceive. Should the believers in special creations consider it unfair thus to call upon them to describe how special creations take place, I reply that this is far less than they demand from the supporters of the Development Hypothesis. They are merely asked to point out a conceivable mode.

a.j.baaqail · 20 January 2010

Ninety five percent of comments of critics of Stephen Meyer's book (signature in the cell) are having same wording like: ' Repetitive junk - Creationist argument - argument from ignorance - making castle from the sand "etc. Permit me to ask those scholars and experts on evolution: What are you people giving us NEW since 150 years on natural selection? Except cursing, naming and denying us. I will tell you something: What is going to happen to the Darwinism, say within 10-15 years? The theory will disappear like other several theories like ' Socialism - Communism-Marxism ' because your Charlie built a hundred storey building, one hundred fifty year ago, but forget to make its foundation and ground floor!

AJ

Dave Lovell · 20 January 2010

a.j.baaqail said: Permit me to ask those scholars and experts on evolution: What are you people giving us NEW since 150 years on natural selection?
How long have you got, and are you prepared to listen? What are you giving us NEW since 6000 years on creationism?
and also: What is going to happen to the Darwinism, say within 10-15 years? The theory will disappear like other several theories like‘ Socialism - Communism-Marxism ‘ .......
Can you promise us that? When are you personally going to stop using a term with no scientific meaning? MET is the best current understanding of how life developed on the this planet, not a political manifesto or moral code.

Dave Luckett · 20 January 2010

I wouldn't want to put you to the effort of actually thinking about this, AJ, but could it be that Meyer actually is making the same false argument over and over, that he really is saying that if he doesn't know how it happened, it couldn't have happened, and maybe it's possible that his entire premise is based on sand, just like the experts say? I mean, it's a wild idea, but it could just be right.

And something new in the last 150 years, you ask? Like, oh, the 4.5 billion year history of the Earth demonstrated and documented, you mean? Or population genetics? Or DNA? Or about a dozen observed speciations? Or radiographic dating? Or the K-T event? Or maybe a hundred transitionals that Darwin never heard of? Or at least five or six different hominid species that aren't modern humans, and aren't modern apes either, being bipedal, with larger brains and different teeth. Stuff like that, you mean?

Foundations and ground floor? Being science, they're made of evidence, AJ. Maybe you've heard the word? Now all you have to do is to work out what it means.

Oh, and one more thing. The Theory of Evolution isn't going to disappear. It's too well-established now - established by the evidence of which you are so sadly and shamefully ignorant. Sadly, ignorance won't disappear either, so long as there are rogues like Meyer to promote it to fools. But don't expect us to do anything but snicker at you when you show it.

DS · 20 January 2010

a.j.baaqail said: Ninety five percent of comments of critics of Stephen Meyer's book (signature in the cell) are having same wording like: ' Repetitive junk - Creationist argument - argument from ignorance - making castle from the sand "etc. Permit me to ask those scholars and experts on evolution: What are you people giving us NEW since 150 years on natural selection? Except cursing, naming and denying us. I will tell you something: What is going to happen to the Darwinism, say within 10-15 years? The theory will disappear like other several theories like ' Socialism - Communism-Marxism ' because your Charlie built a hundred storey building, one hundred fifty year ago, but forget to make its foundation and ground floor! AJ
Well A. J. let's see, how about a short list just off the top of my head: Punctuated equilibrium Neutral theory Hox genes Evo Devo Interference RNA Transposition SINE insertions Cladistics Well, you get the idea. The list goes on and on. In fact, over one million publications is peer reviewed journals contain quite a bit of new information, even entirely new fields of study. Your turn. Name one new idea in creationism in the last one hundred and fifty years. Not made up worthless crap like "complex specified information", not tired old recycled "intelligent design" nonsense, something actually new. See Meyer is just making the same tired old argument over and over, no one buys it. As for the demise of evolutionary theory, all I have to say is WATERLOO, WATERLOO!!! Just keep screaming it until you can't talk anymore. Evolution will still be around long after humans have evolved enough to realize that it is true.

Stanton · 20 January 2010

DS said: As for the demise of evolutionary theory, all I have to say is WATERLOO, WATERLOO!!! Just keep screaming it until you can't talk anymore. Evolution will still be around long after humans have evolved enough to realize that it is true.
People have been screaming this about evolution for the last 149 years. One would think that these people would realize that making false predictions won't change the future. But...

DS · 20 January 2010

P.S.

Let's not forget such favorites as:

Endosymbiosis

Lateral gene transfer

Molecular population genetics

RNA World Hypothesis

See A.J. the thing is that ignorance of every major discovery in the last one hundred and fifty years is not evidence against evolution. It is only evidence against ignorance.

Still waitin for your list lad.

stevaroni · 20 January 2010

a.j.baaqail said: Ninety five percent of comments of critics of Stephen Meyer's book (signature in the cell) are having same wording like: ' Repetitive junk - Creationist argument - argument from ignorance - making castle from the sand "etc.
Alright. Here's a new argument. Meyer spends hundreds of pages talking about the "specified information content" of the genome and how it's so meaningful and important, but in all that time never defines his terms and never tells us how to go measure it. Now, why might that be, AJ? Why has ID, after 30 years of "research" never told us how to actually go measure the most important variable in the ID universe? I would note that science, on the other hand has used the intervening 30 years to years, to decode and publish the entire human genome, and the entire genomes of hundreds of creatures, and to more or less double the number of known transitional forms for dinosaur-to-bird, land-animal-to-whale, fish to-to-amphibian and, of course, ape-to-man.

stevaroni · 20 January 2010

a.j.baaqail said: Ninety five percent of comments of critics of Stephen Meyer's book (signature in the cell) are having same wording like: ' Repetitive junk - Creationist argument - argument from ignorance - making castle from the sand "etc.
Yeah, and ninety five percent of the comments of critics of the TV show "The Bachelor" have some wording like "banal crap". So? When a huge majority of serious reviewers call your product crap it should tell you something.

DS · 20 January 2010

Still waitin A.J. You are getting further and further behind here. Do you really want me to post another hundred examples before you can think of one?

Don't ask questions you don't want answered dude.

Henry J · 20 January 2010

SINE insertions

That won't help; the guy will still go off on a tangent.

Mark Moses · 21 February 2010

It seems to me that Dr. Robert Shapiro's (certainly not a friend of ID theory) stinging criticism of pre-biotic synthesis, upon which the entire RNA First scenario depends, is right on the money. All of these experiments, including those of Joyce and Lincoln, Sutherland, and Szostak, depend on the genius of the experimenter and the advanced technology of the respective laboratories to succeed. To me, this does not offer much insight into how life could have started by itself in a pre-biotic soup 3.6 billion odd years ago. Dr. Graham Cairns Smith and Sir Fred Hoyle were also extremely critical of the techniques of pre-biotic synthesis to help us understand how life started by a purely naturalistic process. At the very least, the sharp criticisms of scientists of this stature should evoke a little bit of humility in discussing the origin of life.

If Stuart Kaufmann can state without reservation that we do not know how life started, I don't why other people who certainly are far below his level of expertise and knowledge cannot say the same thing.

There is currently an "Origin of Life Prize" being offered for a million dollars for someone who can present a "highly plausible" explanation for the genetic material needed for life to exist and evolve. So far nobody has won it. That would mean that, at least in the eyes of the sponsors, there does not exist a highly plausible explanation for the origin of DNA. I don't know if Meyer is right or not about an intelligent designer, but there certainly is nothing unreasonable in considering such an option. I get the feeling from reading some of the comments on this site, that people are afraid of such a possibility.

I also do not understand the "junk DNA" argument. Everyone agrees that DNA is a highly sophisticated digitally encrypted code which contains the instructions to build a living cell. Szostak himself says that "even the simplest bacterium teems with molecular contraptions that would be the envy of any nanotechnologist".
From everything I have read it seems that what goes on in the simplest cell is enough to make your head swim. It seems to me that the "junk DNA" argument is not that DNA does not exhibit design, but that it exhibits "poor design" or "design that leaves something to be desired", or it is not put together in a way that we would have liked to see it. Toyota screwed up on its cars, but no one would conclude from that fact that the Toyota Prius was not the product of intelligent designers.

It seems to me that an honest person should at the very least give some sort of consideration to Meyer's ideas. In the meantime it seems clear that nobody really knows how life did start.

MattK · 11 March 2010

All of these experiments, including those of Joyce and Lincoln, Sutherland, and Szostak, depend on the genius of the experimenter and the advanced technology of the respective laboratories to succeed.
So, if I interpret your argument correctly, you're saying that if an experiment is clever than it can't have anything to say about natural processes. Have I got that right? Fascinating. Btw, citing Fred Hoyle is not exactly convincing.

Roger · 1 June 2010

Please send subscriptions