Ceci n'est pas un blog
Complexity and design
The basic argument of anti-evolutionism (Intelligent Design, ID) is that chance processes cannot account for the complexity that we see in living things. The probabilities are simply vanishingly small.
That may be true, but I'm nervous about that kind of argument. The reason is that I'm not sure we properly understand complexity. There are examples of complex systems that behave in apparently extremely improbable ways, if we calculate the probabilities in a straightforward manner. Complex systems are well-nigh impossible to model, so I'm very uneasy about saying, "The probability that a complex system will do X is 0.0000000000000001."
Here's an example of complex biological systems doing extremely "improbable" things, courtesy of Ard Louis and BioLogos:
| Print article | This entry was posted by Anthony on 24 Nov 2010 at 9.57 am, and is filed under Science & Faith. Follow any responses to this post through RSS 2.0. You can leave a response or trackback from your own site. |
No trackbacks yet.
An open letter to the British Humanist Association
about 1 year ago - 10 comments
Dear BHA, After re-loading your website home page a few times, I found on it the following quote by Ariane Sherine: "All children should be free to grow up in a world where they are allowed to question, doubt, think freely, and reach their own conclusions about what they believe." As an evangelical Christian, I wholeheartedly…
Intelligent Design: a Bayesian summary
about 2 years ago - 5 comments
This post concludes my recent splurge on Intelligent Design (ID). I have been trying to argue: That ID is basically anti-evolutionism That, as anti-evolutionism, ID is half an argument for creationism That the question of whether ID is science is actually quite dull That ID proponents are right to point out the speculative nature of…
Should Intelligent Design be taught in schools?
about 2 years ago - 12 comments
I don't see why not. Many people in the UK believe in Intelligent Design (that is, they don't believe in modern evolutionary theory), 51% according to one report. This in itself is a good reason to expect children at least to be familiar with the term and what it means. They should learn about some…
Can science point to an intelligent designer?
about 2 years ago - No comments
I find myself in full agreement with the basic presupposition of Intelligent Design (ID): that a world in which an intelligent agent has acted might be expected to be different to a world in which no such intelligent agent has acted, and that the methods of science (i.e., empirical enquiry) might be a useful tool…
Darwinian speculations?
about 2 years ago - 7 comments
One of the quotes Michael Behe showed at Monday's Darwin or Design? evening was the following: "We should reject, as a matter of principle, the substitution of intelligent design for the dialogue of chance and necessity, but we must concede that there are presently no detailed Darwinian accounts of the evolution of any biochemical system,…
Is Intelligent Design science?
about 2 years ago - No comments
Countless hours have been wasted poring over this most uninteresting, pedantic and pointless of questions. Rather than seeking to answer the questions raised by Intelligent Design (ID), certain people seem to think it is much more pressing to determine once and for all in which drawer of the filing cabinet the question should reside. Enough!…
Is Intelligent Design creationism?
about 2 years ago - 18 comments
No. But there is a connection. Intelligent Design (ID), being essentially anti-evolutionism, is half of an argument for creationism. Basic scientific arguments go something like this: You are wrong. I am right. Or, in more detail: Your model does a poor job at explaining the data. My model does a better job at explaining the…
Michael Behe: what is "Intelligent Design"?
about 2 years ago - 4 comments
Last night I called in at Westminster Chapel for Darwin or Design? An Evening with Michael Behe, hosted by Justin Brierley, presenter of Premier Christian Radio's Unbelievable? programme, with support from the new Centre for Intelligent Design. I thought I'd use that as an excuse for a series of blog posts about "Intelligent Design" (ID), as…
Creation or evolution: do we have to choose?
about 2 years ago - 18 comments
This question (also the title of a recent book by Denis Alexander) is one over which evangelical Christians often sharply disagree. Some ("young-earth creationists") maintain that the Bible makes it crystal clear that life, the Universe and everything have had their beginning within the past few thousand years. Others ("theistic evolutionists") are less willing to…
I live in York and I 
Loading ...
about 2 years ago
Think of it this way.
What are the odds that Giraffe necks should be so finely tuned as to be exactly (within a fraction of a millimeter) the right length so as to be able to meet their head?
This is such a vanishingly small probability there must have been some design.
Answer - yes - it's called natural selection.
Giraffes with no heads don't reproduce very effectively.
about 2 years ago
Oh and BTW biology has 4 ways in which irreducibly complex things evolve;
deletion of parts
addition of multiple parts e.g. duplications or merging of two complexes
change of function
gradual modification of parts
or combinations of the above.
If I went around claiming that astronomy has no explanation for something that you were learning about what would you think of me?
Regards,
Psi
about 2 years ago
"If I went around claiming that astronomy has no explanation for something that you were learning about what would you think of me?"
I'd think you were spot on. That's why we do research.
about 2 years ago
No, I mean something that you do have an explanation for.
Something that is completely uncontroversial and well backed up by evidence.
What would you think of me then?
about 2 years ago
There are plenty of uncontroversial things backed up by evidence that we don't have an explanation for.
Maybe it would be quicker if you just make your point
about 2 years ago
I am afraid Ard Louis has not really answered Irreducible Complexity.
The reason that protein-self-assembly is possible is that although the flagellum is Irreducibly Complex with respect to its function, it is certainly not irreducibly complex with respect to its assembly and we can see that just by looking at the parts: 2 neighbouring parts must have complementary surfaces. A flagellum assembles step-by-step under the influence of forces between its parts. The reason these complementary surfaces exist, is that they are preprogrammed by the protein sequence; they are full of information and so appear to have been designed. Ordinary molecules would never assemble into such specific arrangements.
On the other hand, the problem is that Darwinism must assemble by selecting for FUNCTION, and is not pre-programmed. That is why Irreducible Complexity is a such a barrier for it.
The information observed in a finished molecular machine, would reliably indicate design in any circumstance outside of biology. That is still true even though we know the structure is self-assembled, because that requires highly specific (not necessarily unique) properties in the sequence, which requires even more ingenious design than direct assembly would. Information, and thus the design inference, is conserved in some sense.
People need to dig into Dembski's stuff as well as Behe. I think the concept of Conservation of Information is even more powerful and fundamental than Irreducible Complexity, as he gives some powerful demonstrations that even Evolution does not escape it. As for Ard Louis, I wish he would step above these facile put-downs.
about 2 years ago
Information is inversely related to probability btw, so Conservation of Information roughly translates to a Conservation of Probability.
about 2 years ago
Andrew - all good points, but I still get uncomfortable when people try to calculate the probabilities of irreducibly complex machines forming by chance. Maybe I should be more comfortable, but Ard's example makes me want to tread carefully.
about 2 years ago
The point I am making is that claims of irreducible complexity made by ID/Creationists either explicitly claim or imply that biology has no explanation of their existence and that such things can not possibly evolve.
The equivalent might be me claiming that astronomers have no idea how the sun shines.
Biology has explanations for these things. If ID/Creationists can find flaws in these explanations they should tackle them. They don't. They just keep claiming such explanations don't exist.
What would you think os someone repeating the claim that science has no explanation for how the sun shines and then drawing a conclusion from this?
about 2 years ago
Andrew,
Dembski cant even come up with a way to measure information that is accepted by the mathematics community.
They have printed their problems with his claims.
He hasn't addressed them.
Conservation of information doesn't exist.
Google gene/chromosome, genome duplication.
about 2 years ago
"Biology has explanations for these things."
Psi - you evidently waste far more of your life than I do on this issue, so could you respond to the quote I gave at
http://www.anthonysmith.me.uk/2010/11/24/darwinian-speculations/
Was it true in 2001 that there were indeed "no detailed Darwinian accounts of the evolution of any biochemical system"? Is it true now?
about 2 years ago
You are thinking of polyploidy, e.g. in plants. But gene/genome duplication does not increase information; it only duplicates it. If I say "the sky is blue" twice, I have not created or transmitted any more information than if I said it once. This is one of Dawkin's famous errors. All it does is increase the expression of certain genes.
Dembski and others are working on ways to measure information. Try this for an example (not by Dembski). http://tbiomed.com/content/4/1/47
For his work on Conservation of Information,
http://evoinfo.org/publications/
But even if we couldnt, should we give up? Wouldnt that be a science-stopper. At the turn of last century, Boltzmann demonstrated that thermal Entropy was at heart a mathematical, statistical problem and was therefore absolute. Scientists of the time hated the idea because it upset their "scientific" (actually materialistic) notion that the universe was eternal, so they ignored his work. He committed suicide. But he was right, and now everyone knows it.
The Information thing is closely related in theory ...
about 2 years ago
If there are responses to Dembskis papers, I genuinely would like to see them.
about 2 years ago
Anthony, I agree with you that discussions of probability need care, but still, Ard ought to be aware that the design inference not only tracks backwards in the case of self-assembly, but actually increases in magnitude, and he ought to be able to see that he has misapplied the concept of Irreducible Complexity.
The concept of Conservation of Probability is not widely understood and accepted, but I think it is sound, if you and Psi want to discuss it more thoroughly.
about 2 years ago
Hi Andrew,
I am partly referring to polyploidy, but also other known ways in which genetic information is duplicated.
Once a duplicate is present is can mutate without preventing the original gene(s) from churning out their proteins.
I am doing a Life Sciences Degree with the OU and the second level core concept course "uniformity and diversity" has whole chapters on the origin and development of "information" in genomes.
To talk about it being in doubt scientifically is laughable. Scientists do real science using what we know about this all the time. Trying to rustle up confusion and denial of this is hardly science, or an honest pursuit.
Which of he following has the most information;
acgtagacgatagacagatagac
or
gagagctcagacgtagctgctag
Neither you, nor Dembski can tell.
"Information Theory" = up in smoke.
I find it hard to believe you can be that interested in the problems with Dembski's work and haven't even googled them.
If you want a recommendation then perhaps start with this
http://scienceblogs.com/goodmath/
This guy covers Dembski's claims about a design filter.
- - -
Please fire away with your "Conservation of Probablity" concept - this is completely new to me.
How is it different from this "Conservation of Information" melarkey?
Cheers,
Psi
about 2 years ago
Hi Anothy,
You said;
Psi – you evidently waste far more of your life than I do on this issue, so could you respond to the quote I gave at
http://www.anthonysmith.me.uk/2010/11/24/darwinian-speculations/
Was it true in 2001 that there were indeed “no detailed Darwinian accounts of the evolution of any biochemical system”? Is it true now?"
Hey this is your blog I am answering man - don't have a go at my life - you barely know me!
I will answer at your post - but the summary is basically yes and no depending on what "detailed" means.
about 2 years ago
psi gets it. No matter how improbable it might be for any stretch of DNA to have come about by chance mutations, it's still MORE likely than any other explanation we know of. Intelligent design rquires that you have confirmable observation of some OTHER explanation, which you can confirm is intelligent. I.e. "God".
Until you have observed this, Darwinian selection is the only thing you've got.
Also, there's no such thing as "complexity". Creationists mistakenly use that term in place of "familiarity with a known cause." Same applies for "information", as psi elegantly demonstrated.
Creationism ala Behe et al is a deluge if statistical fallacies.
about 2 years ago
"gene/genome duplication does not increase information; it only duplicates it. If I say “the sky is blue” twice, I have not created or transmitted any more information than if I said it once."
Oh... WOW. Where to begin.
Gene duplication can cause e.g. extra limbs, which can increase an organism's probability of replicating.
And once the duplication has happened, those two segments are free to mutate independently, which can lead to e.g. the difference between legs and arms.
about 2 years ago
psi,
You say "biology has 4 ways in which irreducibly complex things evolve;"
While that's academically interesting, it's completely irrelevant to this "debate". Even if scientists generously assume, for the sake of argument, that creationists are correct that an given system is "irreducible", all that means is that random mutations would have been less likely to produce it.
But since random mutations are THE ONLY KNOWN EXPLANATION WE HAVE, they are still the best explanation.
Until we observe some other phenomenon which can account for arranging DNA in such a way as to produce the "irreducible" system, that's the state of things. And even if we do see some other phenomenon, we would have to confirm that it is intelligent before we could then conclude "intelligent design".