Stephen Meyer's Bogus Information Theory
A couple of months ago, I finished a first reading of Stephen Meyer's new book, Signature in the Cell. It was very slow going because there is so much wrong with it, and I tried to take notes on everything that struck me.
Two things struck me as I read it: first, its essential dishonesty, and second, Meyer's significant misunderstandings of information theory. I'll devote a post to the book's many mispresentations another day, and concentrate on information theory today. I'm not a biologist, so I'll leave a detailed discussion of what's wrong with his biology to others.
In Signature in the Cell, Meyer talks about three different kinds of information: Shannon information, Kolmogorov information, and a third kind that has been invented by ID creationists and has no coherent definition. I'll call the third kind "creationist information".
101 Comments
Glen Davidson · 13 January 2010
http://tinyurl.com/mxaa3p
Jeffrey Shallit · 13 January 2010
Yes, somehow creationists are always concerned with evidence for design, but never seriously consider how our examples of human design differ in many ways from what we see in biology. In particular, they don't seem to have seriously addressed Mark Isaak's article.
eric · 13 January 2010
I note Meyer also has a tendency (in those quotes) to use the weasel word 'large'. I guess whenever a biologist demonstrates evolution has produced useful information, he'll just claim the amount isn't large enough to count as a violation of his rules.
The focus on "large amounts of specified information" (my emphasis) also reduces his entire book to a strawman argument, since the TOE neither predicts nor expects saltation. It is the compounding of small changes across generations that makes net change in the genome large. If "small" amounts of specified information can be produced naturally under Meyer's (false and arbitrary) limits, this allows evolution and speciation to occur. In other words, you could believe what he says about information and he'd still be wrong about design being necessary in biology.
Matt Young · 13 January 2010
John Kwok · 13 January 2010
John Kwok · 13 January 2010
Jeffrey -
Thanks for yet another insightful post. Maybe you can enlighten BioLogos's Darrel Falk, who contends that there is some utility in Intelligent Design based on its interest in biological "information". Even after I pointed out that you, among others, have rejected the "importance" of this work, he still believes that there is this utility.
yum install Jesus · 13 January 2010
MACROEVOLUTIONISTS THINK ALL THINGS HAPPEN IN RANDOM EXPLOSIONS. IT IS NOT TRUE. THE VOLCANO VESUVIUS DESTROYED THE CITY OF POMPEII; IT DID NOT BUILD A BETTER CITY ON TOP OF IT. INFORMATION THEORY PROVES CITIES HAVE TOO MUCH COMPLEX SPECIFIED INFORMATION TO EMERGE WITHOUT INTELLIGENT DESIGN, AND EVEN A PROKARYOTIC CELL HAS MORE BITS OF COMPLEX SPECIFIED INFORMATION THAN A CITY, AND A EUKARYOTIC CELL HAS EVEN MORE. THE DARWINIACS HAVE YET TO SHOW ONE CITY THAT WAS BUILT BY A VOLCANO, AVALANCHE, OR OTHER NATURAL DISASTER. UNTIL THEY DO, THEIR THEORY REMAINS UNPROVED!
Matt Young · 13 January 2010
John Kwok · 13 January 2010
Jeffrey Shallit · 13 January 2010
Jesus:
Information theory, even creationist information theory, has nothing to do with "building a better city". The notion of "better" doesn't exist in information theory.
Explosions are a poor analogy with mutation+selection.
And finally, your argument is not improved by shouting.
DS · 13 January 2010
YUM YUM THINK ALL THINGS HAPPEN IN NON-RANDOM MIRACLES . IT IS NOT TRUE. THE VOLCANO VESUVIUS DESTROYED THE CITY OF POMPEII; IT DID NOT INVOLVE ANY PUNISHMENT FROM GOD. INFORMATION THEORY PROVES THAT COMPLEX SPECIFIED INFORMATION IS PURE CRAP AND SO IS INTELLIGENT DESIGN, AND EVEN A PROKARYOTIC CELL HAS NO BS INFORMATION, AND A EUKARYOTIC CELL HAS SOME BUT WE CANT SAY HOW MUCH THAT COULD ALL BE PRODUCED BY NATURAL MEANS. THE CREATIONISTS HAVE YET TO SHOW ONE THING THAT WAS BUILT BY A MIRACLE, OR OTHER NON-NATURAL PROCESS. UNTIL THEY DO, THEIR SO CALLED THEORY REMAINS COMPLETELY UNSCIENTIFIC!
stevaroni · 13 January 2010
YUM: ON THE LEFT EDGE OF YOUR KEYBOARD THERE IS A KEY NAMED "CAPS LOCK". IT"S PURPOSE IS TO MAKE YOUR RANTS READABLE.
USE IT (Caps) like this. Now your rants will be in sentence case, which is actually readable, if not any more sensical.
Additionally, on the right edge of your keyboard is an L-shaped key labeled "Enter".
You use this key to make things called "paragraphs". Paragraphs separate things called "thoughts" and again, make your rant more readable.
You use it like this (enter)(enter)
First nonsensical thought (enter)(enter)
Initial non-sequitor (enter)(enter)
Long list of subsequent non-sequitors (enter)(enter)
See?
harold · 13 January 2010
I had the thought the other day that there is a central flaw in all creationist claims about information.
In all legitimate information theories, information is implicitly created by the observer. Something that is "noise" to me may be information to you.
Creationists always claim that intelligence must be created for the observer. A magic "designing" "intelligence" had to create something before it became information.
This quickly leads to absurd conclusions. If I decided to measure the weights of individual grains of sand from a sample for some reason, that distribution becomes information from my observer perspective. Therefore, by "creationist information theory", the grains of sand must have been "designed by an intelligence".
Dan · 13 January 2010
RDK · 13 January 2010
Shebardigan · 13 January 2010
R0b · 13 January 2010
Thank you, Dr. Shallit. I was interested to see a poster on Telic Thoughts lay down a challenge to "go where reviewers are reluctant to trod", namely to point out specific problems in Meyer's facts and reasoning, and to back up our criticisms. I submit that anyone who considers this task to be challenging has not read the book very carefully.
A real challenge would be to point out all of the problems in the book. I've been working on chapters 4, 8, 13, and 16, as time permits, and it's looking to take quite a while.
fnxtr · 13 January 2010
As I said on your blog, JS:
p. 258 "If a process is orderly enough to be described by a law, it does not, by definition, produce events complex enough to convey information."
Somebody call Johannes Kepler, there's no way to know where the planets are!!!ONE!!!
John Kwok · 13 January 2010
J-Dog · 13 January 2010
RDK - I too once thought that the Jeebus yum was a Poe, but if you go to his website, you will see a virtual self-spammed site reeking of flat-earthism and other freakish stuff.
If it's a Poe, it's a damn good one, ONE THAT SHOULD BE POSTING OFTEN AT UD! (That is Dembski's Uncommon Descent site for the unitiated). IF I WERE A JEEBUS YUM, OR EVEN A POE, I WOULD GET OVER ASAP TO UD AND SAVE SOME DAMN SOULS!
DO IT FOPR THE KIDZ.
Matt G · 13 January 2010
Dan · 13 January 2010
Mike Elzinga · 13 January 2010
Henry J · 13 January 2010
DS · 13 January 2010
DS · 13 January 2010
DavidK · 13 January 2010
Excellent review. If still available I'll try to reference it on the Amazon site.
The arguments used by Meyer are understandable in the sense that they appear logical, rational, and scientific. However, considering the fact that Meyer's audience is the fundamentalist churchies who eat up this crap and have NO understanding of science, it all sounds so plausible to them. Thus the Dishonesty Institute is making headway on their wedgies by duping their flocks into this error filled book, sorry, "book of the century" I think some reviewer called it.
Glen Davidson · 13 January 2010
http://tinyurl.com/mxaa3p
Tex · 13 January 2010
DS · 13 January 2010
Nagal wrote:
"But the RNA world hypothesis Fletcher offers as a refutation is carefully described by Meyer, who argues that while it might help solve some problems (in virtue of the catalytic properties of RNA), it simply pushes back to a different molecule the basic question of how such an extremely complex replicator came into existence, thus allowing natural selection to begin."
Actually that's not a very big problem at all. First, the precursors of ribonucleotides are some of the most likely molecules to form under the conditions of the primitive earth. Second, ribonucleiotides can be polymerized by chemical reaction catalyzed by things other than proteins. Third, RNA molecules can serve as templates for their own replication and can catalyze their own replication. Taken together, these observations mean that it extremely likely that such a replicating system could arise and undergo selection, given the conditions of the primitive earth. It certainly doesn't solve all the problems, but it does go a long way.
John Kwok · 13 January 2010
@ Glen -
Speaking of dishonesty, you may include the likes of Karl Giberson and Darrel Falk. Neither one believes that the Dishonesty Institute has sinister motives (In fact Giberson says he knows some at the DI and that they are a-ok with him, except that they are pushing ID as a political, instead of scientific, movement.). Falk thinks that there is some potential usefulness from the DI with regards to our understanding of "biological information" (Of course that's nonsensical, given the substantial critiques written by Jeffrey Shallit and Wesley Elsberry, among others.).
They are also dishonest in the sense that they allow creos to post ad hominem attacks, but when I have sought to remind them of some uncomfortable truths regarding the DI's current and past behavior, I've been told to "shut up". To make a long story short, I've been banned there until I "apologize" for my behavior.
At least AusAdrian (#2423, here: http://biologos.org/blog/footprints-in-the-sand/), you and one other frequent PT poster appreciated my efforts in trying to persuade both Giberson and Falk that they are mistaken in thinking that the DI could be persuaded to "play ball" with the mainstream scientific community. In fact, both agreed with my observation that the BioLogos Foundation could be viewed as a DI "stealth ally".
As AusAdrian noted in his comment, all I tried to do there was this: "...I found your writings so refreshing and reinforcing." Sometimes the truth hurts and it seems as though BioLogos isn't interested in hearing the truth about the ongoing lies, censorship and other forms of deceit, including theft, being practiced by the DI daily.
Sincerely,
John
Frank J · 14 January 2010
Stanton · 14 January 2010
Karen S. · 14 January 2010
harold · 14 January 2010
Peter Henderson · 14 January 2010
harold · 14 January 2010
Peter Henderson -
In my view, the creationist use of "information theory" is exactly equivalent to the creationist use of "the second law of thermodynamics".
They are simply trying to change the subject to something that they desperately (and wrongly) hope that biologists are unfamiliar with - but that they themselves are completely ignorant of - and do an end run around the evidence, by (wrongly) claiming that evolution is "theoretically impossible" (so all of the evidence for it can't matter).
I'm just an MD pathologist (no longer practicing) with an undergraduate degree in biology and some master's level courses in statistics and probability, not an information theorist.
Yet as I noted above, the creationist approach seems to me to be grounded in a simple, fundamental error. Information is created by the observer, that's trivially true and compatible with all valid information theories. But all creationist claims amount to the assertion that it had to have been created "for" the observer - that if an observer makes something into information, that something had to have been directly "created by intelligence" in the first place. As I also noted, this leads to absurd conclusions, because any arbitrary phenomenon will become information if observed in an organized way by an intelligent observer.
But of course, they'll make up nonsense arguments and contradict actual experts like Jeffrey Shallit all day.
harold · 14 January 2010
Peter Henderson -
Whoops, my fourth paragraph above borders on an unintentional straw man misrepresentation of actual creationist claims. Let me immediately clear that up.
What they actually assert is that certain "kinds" of information had to be directly "created by intelligence". "Complex specified information" and so on.
The problem is that the creationist invented terms have no meaning that creationists themselves cannot define them in a meaningful way.
Jeffrey Shallit · 14 January 2010
If someone claims It may be a word in each of those languages but a word does not constitute information, then there is really nothing to do except roll one's eyes. In both Shannon's theory and Kolmogorov's theory, a word or string of symbols indeed constitutes information. To deny this just exhibits utter cluelessness.
eric · 14 January 2010
Jeffrey Shallit · 14 January 2010
Peter: I'm in agreement. It's just fine for Dembski or anyone else to propose a new definition of information. But then he should (a) publish it in the peer-reviewed scientific literature (b) forthrightly deal with criticisms (c) acknowledge mistakes quickly and (d) show how his ideas can be used on more than toy examples. If only he had done all these things, instead of heaping scorn on his critics, calling them "obsessive" and worse.
One of the very first things Elsberry and I commented on was the observer-dependence of "specified complexity".
Frank J · 14 January 2010
eric · 14 January 2010
Thinking about why we have two different, incompatible definitions of information can also give insight into why no actual scientist thinks much of Meyer's version.
Why do we have two? Why not figure out which one is better and just use that one?
Well, the answer is we have two because both are useful. Shannon is good for answering some questions, solving some practical, real-world problems. Kol for others.
So what would make scientists adopt a third definition? Why, that's easy: utility! Armed with this insight we can respond to the YECers. There is no algoritm or rules set for determining how to measure specified information. One scientist cannot check another's results, because they can't even figure out how to measure it in the first place. Because of that, its useless. And because it is currently useless, science has not adopted Meyer's idea.
In a sense the argument over specified information reflects a broader difference between creationists and scientists. To creationists, a theories' reflection of Truth is more important than its utility. Specified complexity seems to reflect an idea of platonic, non-material, objective information. Thus creationists view it as a good theory, even if its useless for (e.g.) helping you compress files. In contrast, scientists value utility far more than how well a theory matches some preconceived notion of Truth. Shannon and Kolomgorov are useful to science. Meyer is not. So Shannon and Kol will continue to be taught in information theory, while Meyer will not.
If Meyer or Dembski ever give us a methodology for calculating specified information which can be used independently, and if this methodology turns out to be useful, then it will get respect. Not before.
phantomreader42 · 14 January 2010
Scott · 14 January 2010
jerrym · 14 January 2010
eric · 14 January 2010
harold · 14 January 2010
John Kwok · 14 January 2010
John Kwok · 14 January 2010
Karen S. -
I don't think they've backed off the "stealth ally" comment. They opted to delete my list of Dembski's "crimes", allowing me to keep it only at one blog thread (I had posted it more than once, the second time to remind Karl Giberson that Dembski is not a nice person, nor are his fellow Dishonesty Institute mendacious intellectual pornographers.).
Karen S. · 14 January 2010
John,
I think one of them really did back away somewhat from the comment. As I recall, Peter Enns asked you to document the DI's dishonesty, which you did. Any feedback from that? Why not post a link to Jeffrey Shalitt's latest post, which is specifically on their dishonesty?
John Kwok · 14 January 2010
Karen S. · 14 January 2010
John,
Are you kidding? Do they make anyone else apologize? How do they want you to apologize-- via email or on the blog? There seems to be lots of whack jobs commenting on that blog and many are very insulting, e.g. Gregory. When John Davison was posting there he called me a bitch! Glen Davidson was the only one who told him off for that.
Anyway, the blog doesn't appear to be friendly at all to women. The bloggers don't respond to my comments except for 1x only. There are no female bloggers and very few female commenters.
Scott · 15 January 2010
John Kwok · 15 January 2010
ben · 15 January 2010
John Kwok · 15 January 2010
John Kwok · 15 January 2010
Karen S.,
In light of my recent experience over at BioLogos, I must conclude that they are more interested in winning the "approval" of their fellow "brothers in Christ" than in reminding them that they must recognize the scientific fact of biological evolution and that Intelligent Design doesn't even come close as a "substitute" to contemporary evolutionary theory. Their stance is substantially at odds with invertebrate paleontologist Keith Miller (an Evangelical Protestant Christian) or Ken Miller's (who, I will observe again, did tell a private audience here in New York City last May that those who belong to faiths hostile to science should leave such faiths for those which do accept science).
harold · 15 January 2010
Scott -
Well, not quite exactly, but certainly close enough. We both accept standard, rigorous definitions of information, and agree about creationist mistakes.
The source I quoted does actually say that something definitively becomes information upon being observed as information. That position is not at all controversial.
However, you're obviously correct that, in most contexts, under the assumptions of science, the physical entity existed before being observed.
I'd say that the position I quoted is a bit more generalizable than your position, and that you add something that, while obviously true much of the time, may be unnecessary.
Although the position I quoted may superficially sound less concrete, it's actually more concrete and generalizable. There may be disputes about whether or not some type of information fully existed as information before observation, but we can ALWAYS agree that it's information once a human observer makes it so in a manner consistent with information theory.
However, enough of this, because we clearly don't have a real disagreement, and creationist distortions and dissemblings grossly exceed this kind of philosophical fine tuning.
As for creationists, as I noted above, they use "information theory" in the same way that they have classically used "the second law of thermodynamics" (and many other things).
Frustrated by trying to dispute the evidence evolution, they try to do an end run around it by pulling a "theoretical disproof" of evolution out of their butts. Note the absurdity. A one-liner "theoretical disproof" of a strongly supported theory must always be viewed with great skepticism.
And in the course of trying to do this end run, they are by necessity required to try to change the subject.
They desperately (and incorrectly) hope that biologists will be extremely ignorant of some other field (thermodynamics, information theory, whatever), and that full experts in that field will be unaware of their claims about it.
(However, they get one thing right - they assume correctly that the dolts who actually give them real money will be as ignorant of [physics, information theory, or whatever] as they are of biology.)
They then proceed to distort and dissemble on (thermodynamics, information theory, whatever), revealing their own ignorance and chicanery.
And inevitably they get busted by biologists and full experts in the field that they are attempting to distort.
harold · 15 January 2010
John Kwok -
I'm inclined to agree with Kenneth Miller on the matter of religions that deny scientific reality, with the strong caveat that other peoples' religious beliefs are not my business unless they want my opinion, unless they do something that affects me.
It seems to me that getting emotionally committed to a position that outright denies scientific reality (not a position that believes things that are beyond the realm of science, or beyond the current reach of scientific investigation, but a position that denies scientific reality), inevitably leads to some frustration.
Creationists are a good example of this.
Rilke's granddaughter · 15 January 2010
Mike Elzinga · 15 January 2010
DS · 15 January 2010
Mike wrote:
"For the ID/creationist, the process of decay after a living organism dies is their classic example of what they misrepresent as the second law of thermodynamics in action."
Great, now we can agree that dead organisms cannot evolve. Finally, some common ground!
John Kwok · 15 January 2010
Apparently RG, you're the only one who doesn't care, since you've decided to be a "good" troll and stop by to comment. A few others have commented on it, most notably Karen S.
It's hysterical that you seem to "drop in" whenever I say something that strikes you as being funny or silly or outrageous. How come you don't "visit" when I truly have something commendable?
Do me a favor. If you don't like my posts, then stop "dropping by" to comment, please.
John Kwok · 15 January 2010
John Kwok · 15 January 2010
Rilke's granddaughter · 15 January 2010
John Kwok · 15 January 2010
nmgirl · 15 January 2010
Karen S. · 15 January 2010
I guess if you don't like John's comments you don't have to read them.
Rilke's granddaughter · 15 January 2010
Rilke's granddaughter · 15 January 2010
Stanton · 16 January 2010
Rolf Aalberg · 16 January 2010
Rolf Aalberg · 16 January 2010
than dammit
John Kwok · 16 January 2010
Thank goodness I have a life which extends far beyond responding to self-centered, quite arrogant, trolls like yourself (For example, I am about to do my small bit in helping relief efforts in Haiti. When you can do something like that, then and only then, do you have any right to judge someone other than yourself.).
John Kwok · 16 January 2010
Stanton · 16 January 2010
Karen S. · 16 January 2010
Additionally, I happen to like Star Trek references. But I'm wondering-- if Spock descended from both humans and Vulcans, why are there still humans and Vulcans?
John Kwok · 16 January 2010
Ichthyic · 17 January 2010
Why don't you "get over yourself," RG. Lately it seems as though half the posts you do post here pertain me.
maybe it's because often half the posts here ARE yours?
you have NPD, Kwok.
seek treatment.
Rilke's granddaughter · 17 January 2010
Dale Husband · 17 January 2010
John Kwok · 17 January 2010
Rilke's granddaughter · 17 January 2010
Dale, I normally leave John alone. But he's a cyberstalker of women, and that bears watching. Do you enjoy him constantly whining about his latest blog-banning?
Dale Husband · 17 January 2010
John Kwok · 17 January 2010
John Kwok · 17 January 2010
John Kwok · 17 January 2010
Rilke's granddaughter · 17 January 2010
Rilke's granddaughter · 17 January 2010
Rolf Aalberg · 18 January 2010
I agree John could write less about things of interest to very few people beside himself. OTOH, I cant see RG's comments add much of value to this thread either. Although I don't know about other threads.
Why not use e-mail?
fnxtr · 18 January 2010
So anyway, about Myers' so-called theory... a book that contains, as DA would say, a "strange usage of the word information I wasn't previously aware of", actually contains no new information whatsoever.
There's one for Alanis.
fnxtr · 18 January 2010
Oh dear. Meyer's. Sorry, PZ.
eq · 31 May 2010
Did any of you even read Dr. Meyer's book?? He addresses all of the above issues in detail. Try reading the book then get back to me for some serious debate!
MrG · 31 May 2010
Mike Elzinga · 31 May 2010
MrG · 31 May 2010
Aw, you're too serious MrE. I was amused by the "serious debate" remark. Every time I hear a creationist say something along such lines I back up, since it's likely to be followed by the heel of his boot planted firmly midway between my hip joints.
Mike Elzinga · 31 May 2010
Stanton · 31 May 2010
GODDESIGNER" is supposed to engender "serious debate." As far as I've seen, no Intelligent Design proponent is interested in "serious debate," whether concerning Intelligent Design, or science in general.Malchus · 31 May 2010