Dembski’s Latest: “Life’s Conservation Law”, and why it’s stupid

So. William Dembski, the supposed “Isaac Newton of Information Theory” has a new paper out with co-author Robert Marks. Since I’ve written about Dembski’s bad IT numerous times in
the past, I’ve been getting email from readers wanting me to comment on this latest piece of intellectual excreta.

I can sum up my initial reaction to the paper in three words: “same old rubbish”. There’s
really nothing new here – this is just another rehash of the same bankrupt arguments that Dembski has been peddling for years. But after thinking about it for a while, I realized that Dembski
has actually accomplished something with this paper: in his attempt to argue that
evolution can’t possibly outperform random-walks without cheating, he’s actually explained exactly how evolution works. He attempts to characterize that as cheating, but it doesn’t work.

Let me start with a quick review. Dembski has, for years, been pushing an argument based
on some work called the No Free Lunch (NFL) theorems. The NFL theorems prove that
average over all possible search landscapes, no search algorithm can outperform a random
walk. The NFL theorems are true and correct – they’re valid math, and they’re even
useful in the right setting. In fact, if you really think about it, they’re actually quite
obvious. Dembski has been trying to apply the NFL theorems to evolution: his basic argument
is that evolution (as a search) can’t possibly produce anything without being guided by
a supernatural designer – because if there wasn’t some sort of cheating going on in the evolutionary search, according to NFL, evolution shouldn’t work any better than random walk – meaning that it’s as likely for humans to evolve as it is for them to spring fully formed out of the ether.

Thes doesn’t work for a very simple reason: evolution doesn’t have to work in all possible landscapes. Dembski always sidesteps that issue.

Let me pull out a metaphor to demonstrate the problem. You can view the generation of
a notation for a real number as a search process. Suppose you’re given π. You first
see that it’s close to 3. So the first guess is 3. Then you search further, and get closer – 3.14.
That’s not quite right. So you look some more, and get 3.141593. You’ll get closer and closer
to a notation that precisely represents π. Of course, for π, you’ll never get to an
optimum value in decimal notation; but your search will get progressively closer and closer.

Unfortunately, most real numbers are undescribable. There is no notation that
accurately represents them. The numbers that we can represent in any notation are a miniscule
subset of the set of all real numbers. In fact, you can prove this using NFL.

If you took Dembski’s argument, and applied it to numbers, you’d be arguing that because most
numbers can’t be represented by any notation, that means that you can’t write rational numbers
without supernatural intervention. Of course, that’s rubbish. The way that we create notations for
rational numbers works, because natural numbers are a small, structured subset of the set of all
real numbers. We don’t expect them to work for all possible numbers, and any proof about notations
that relies on reasoning about how well a notation works on all real numbers is absolutely
irrelevant to a discussion of notations for the rationals.

The same thing is true of evolution. Ignoring all of the (very serious) problems with modeling evolution as a search process, evolution doesn’t need to work in all possible search spaces; it needs to work in one particular set of search spaces that by their nature have a lot of structure. Evolutionary processes exploit that structure.

Now, after that downright Oracian introduction, let’s get to Dembski’s latest paper.

It gets off to a very bad start. It starts with a section titled “The Creation of Information”. This is, frankly, a muddled mess. Unfortunately, I think that that is deliberate. You see, Dembski uses very peculiar definitions of information; or, to be more precise, he
doesn’t use any consistent definition. He pretends to use Shannon information theory, but
he tends to vacillate between the Shannon formulation; his own mangled probabilistic formulation;
and Kolmogorov-Chatin. In this section, he purports to talk about creating information – but
what he actually does is try to muddy things up by mixing up a tiny bit of Shannon theory with a variety of non-mathematical philosophers talking about the meaning of information. That’s
not a particular useful endeavor: what Chesterton meant when he talked about acts of will as
self-limitation is not the same thing as what Shannon meant when he said that
information is the elimination of possibilities. The whole section is just at attempt to confuse
things by equating the mathematical definition of information in Shannon theory with
philosophical definitions of information. The problem is that philosophers define information
in terms of meaning; mathematical information theory doesn’t give a rats ass about meaning: a spinning neutron start generates more information (without any intrinsic meaning) in a second than Dembski’s meaningful actions will in his entire lifetime.

That leads directly into section two: “Biology’s Information Problem” – a thoroughly
redundant section of the paper. Whether his argument makes sense or not, this is just
a rehash of what he said in the previous section, and what he’ll repeat in later sections. “Life can’t create information without intelligence, because information doesn’t exist without
intelligence, because according to my conflation of mathematical information and philosophical meaning, it makes no sense to talk about information absent intelligence”. Of course, when it comes down to it, that’s Dembski’s entire argument: by definition information requires intelligence; therefore if living things contain information, it must have been created by intelligence. (If I were Michael Egnor, I’d just wave my hands and say that it’s just a tautology,
therefore it’s meaningless, and be done with it.)

Section three is Dembski once again rehashing hiss argument against Dawkins’ “weasel” example.
The selection function in “weasel” knows the target, and therefore it’s cheating by “smuggling”
information into the search, and evolution couldn’t do it without cheating. God but I’m tiring of
his repeating that same idiotic argument. It’s a tiny, silly, throwaway example that demonstrates
one minor feature of how evolutionary processes work; it was never intended to be a complete
example of how evolution produces life. Based on his refutation of Dawkins, Dembski concludes
“Evolution, despite Dawkins’s denials, is therefore a targeted search after all.” Sorry Bill, but that’s bullshit, and you know it.

Section four: “Computational vs. Biological Evolution” gets even worse. It’s Dembski’s attempt
to pull his “Universal Probability Bound” into the picture. The UPB is one of Dembski’s dumbest
ideas: it’s an argument that there’s some probability threshold where anything less probably is
absolutely impossible. (Of course, he doesn’t want to admit that it’s his own stupid
argument – so he attributes it to Seth Lloyd. And of course, what Lloyd actually said is quite
different from what Dembski tries to imply that he meant.) What’s sad is just how badly Dembski
builds an argument around this.

Anyway, he wants to build up the argument that you can’t possibly have an evolutionary
“search” produce what we see of life in the entire lifespan of the universe. He starts by talking
about an IBM supercomputer than runs at just over a 1 petaflop, and how large the search space
that it could explore is: if it took one floating point instruction per sample from the search
space, then in the lifespan of the universe, it could have searched 1034 samples.
That’s supposed to impress you – and it’s used in the following paragraphs as part of one of the
typical idiotic probability arguments about life – that’s the fastest computer ever, and it could
only search 1034 samples. But he then goes on to say “It is estimated that 8 t the
total number of organisms, both single-celled and multi-celled, that have existed on the earth
over its duration (circa 4.5 billion years) is m = 1040. Thus it would take a million
Roadrunner supercomputers running the duration of the universe to sample as many “life events” as
have occurred on the earth.”. Wow, way to undermine your own argument, Bill. You’ve just admitted
that a million of the fastest computers ever, running since the creation of the universe, couldn’t
search as much space as the biological history of the earth.

As I mentioned, then he launches into another of those idiotic probability arguments. (It’s like he doesn’t want to miss the chance to include a single stupid argument!):

Most search spaces that come up in the formation of biological complexity are far too
large to be searched exhaustively. Take the search for a very modest protein, one that is, say,
100 amino acids in length (most proteins are several hundreds of amino acids in length). The space
of all possible protein sequences that are 100 amino acids in length has size 20100, or
approximately 1.27×10130, which exceeds Lloyd’s limit. For this space, finding a
particular protein via blind search corresponds to a 1 in 10130 improbability. Exhaustively or blindly searching a space this size to find a target this small is utterly beyond not only present computational capacities but also the computational capacities of the universe as we know it.

Why do we need to see this bullshit time and time again? No one with the slightest shadow of a
clue claims that life requires exactly one possible protein out of all of the incredible
number of possibilities. No one with the slightest shadow of a clue claims that all possible
proteins are equally likely. No one with the slightest shadow of a clue would claim that the
production of any protein is the result of a targeted search for that protein. This is pure
stupidity – a pathetic strawman argument that’s been discredited literally hundreds of times. But
Dembski trots it out and babbles about it, at length.

Finally, in section 5: “Active Information”, Dembski finally gets to the point: that
any search that creates new information must actually have that information somehow included in
the search function. And of course, to make that argument, once again he trots out “Weasel”. Yes,
once again – the argument comes down to “The reason that weasel works is because the selection
function knows its target.” Well, duh. Yeah, Bill? Weasel is a stupid throwaway example. Care to
take on something actually real? Like, say, some of the fantastic e.coli experiments? No,
I didn’t think so.

Then it’s time for obfuscatory mathematics. In case you haven’t seen the term before, it’s
what I call the use of pointless equations that do nothing except look really complicated, and
give you the appearance of having actually done something deep. One of the basic facts of math,
like programming, is that it’s garbage-in, garbage-out. You can derive really incredibly looking
equations using perfectly valid proofs, to demonstrate any conclusion that you want. But your
conclusion is only valid in a setting where you’ve accurately modeled the
reality that you’re trying to describe.

Search is a lousy model for evolution; general search is a particularly lousy model. I’ve discussed it plenty of times before – for example here – but the problems basically come down to a few simple points:

1. As a search, evolution is a multidimensional search. Most of our intuitions about
search landscapes is based on two or three dimensions. But evolution as a landscape
has hundreds or thousands of dimensions; our intuitions don’t work.
2. Evolution is a dynamic landscape – that is, a landscape that changes
in response to the progress of the search. Pretty much every
argument that Dembski makes can be thrown out on the basis of this one fact:
all of his arguments are based on static landscapes. Once the landscape can change,
every single one of his arguments become invalid – none of them work in dynamic
landscapes.
3. As a search, evolution doesn’t have to work on all possible landscapes. It
doesn’t even need to work on most landscapes. It works on landscapes that have
a particular kind of structure. It doesn’t matter whether evolution will work in
every possible landscape – just like it doesn’t matter that fraction notation doesn’t
work for every possible real number. What matters is whether it works in the particular
kind of landscape in which our theory says it works. And on that question, the answer
is quite clear: yes, it works.

Anyway – using his lousy model of evolution as search, he comes up with what he calls the “Law
of conservation of information”. We’ll ignore, for the moment, the fact that it’s not actually a
conservation law at all. What it says, basically, is that if a search algorithm performs better
than blind search by some factor f(I), that algorithm must have been “purchased” at an information cost I.

What does that “purchased” mean? Basically, that the search function must include, in some form, an amount of information about the search space no smaller that I.

Which comes back to one of the whole objections to the whole NFL-based approach: Dembski
insists on searches that perform equally well over all possible landscapes. Saying that a
search algorithm performs only on some set of search spaces is exactly the same as saying
that the search algorithm contains information about those search spaces!

Dembski’s argument comes down to “Where did that information come from?”; and his answer is
that it must have been put there by someone. How do we know that? Well, because this
whole argument just showed that for a process to generate information, that process must
have contained the information to begin with. So because information doesn’t come from
nowhere, it must have been put there. It’s actually a subtly circular argument: we’ve just
gone through this mass of obfuscatory mathematics to derive a set of theorems that
prove that successful searches must in some sense encode information about the landscapes
that they search. But that doesn’t really matter to the argument. Because the
real argument is that only an intelligent agent can create information. The whole
exercise of deriving the so-called conservation of information laws was based on
the premise that you can’t create information; and now after all of that noise, we’ve come
full circle to show that the conservation of information laws can be used to prove that you
can’t create information. The whole argument is ultimately concluding one of its premises.
We can eliminate all of the faux math, and reduce the argument to its simplest form:
“Only an intelligent agent can create information, therefore only an intelligent agent
can create information”.

Back at the beginning of the paper, I said that Dembski actually manages to
basically refute his own argument – that he shows how evolution can actually work. By now, you should see how that happens: this whole argument comes down to asking what it means to drop the “over all possible landscapes” part of NFL. If you do that, then you end up with a search algorithm that can perform very well on some set of landscapes. Which is exactly what us lousy evolutionists have been saying all along.

1. #1 Bayesian Bouffant, FCD
May 7, 2009

evolution doesn’t need to work in all possible search spaces; it needs to work in one particular set of search spaces that by their nature have a lot of structure.

If evolution searched all possible solutions, we would have at least three pairs of limbs: one for walking, one for manipulating tools, and of course a pair of wings for flying. Instead, we find ourselves trapped in the tetrapod bodyplan.

2. #2 Mu
May 7, 2009

Sorry, to be truly Oracian, your intro has to rip at least 5 different people, and contain at least a dozen links to previous posts. But six ain’t bad for starters.
What always gets me at the “100 peptide” search argument is – it assumes nature is forming random stuff, tries it, discards it, and starts over. And that there is only ONE distinctive correct solution. Since most amino acids have a close equivalent that works nearly the same, the search quickly collapses from 20^100 to 5^100, 10^134 to 10^70. Still a huge number, but at least we’re below the infamous particles in the universe threshold.

3. #3 Bayesian Bouffant, FCD
May 7, 2009

The link you provided looks like Marks’ web site. Did Dembski and Marks manage to publish this in a real journal?

4. #4 James F
May 7, 2009

So was this one of the two papers in press by Dembski and Marks that made it past peer review? Except…this isn’t in a peer-reviewed scientific journal. And the next one won’t be, either.

5. #5 Mark C. Chu-Carroll
May 7, 2009

Re #3:

Not as far as I know.

In the bibliography, they list two other of their papers, which are listed as “to appear”, with the journal names black out. If you delayer the PDF to see the text under the blackouts, it says “International Journal of Fun and Games – You’re too Clever – Gotcha!” – which I take both to be a sign of lack of professionalism, and an admission that they can’t get this stuff published anyplace real.

6. #6 Bayesian Bouffant, FCD
May 7, 2009

What always gets me at the “100 peptide” search argument is

A tour of any of the available sequence databases will reveal that thousands of possible sequences can perform the same polypeptide role, and that is just sampling sequences that actually exist in nature in various species. A little more knowledge of biochemistry tells you that sometimes thoroughly different protein sequences & structures can perform nearly the same role (e.g. chymotrypsin and subtilisin and their similar mechanisms for proteolysis). This falls under the banner of convergent evolution.

7. #7 James F
May 7, 2009

If you delayer the PDF to see the text under the blackouts, it says “International Journal of Fun and Games – You’re too Clever – Gotcha!”

Intelligent Design: we do it for the lulz.

8. #8 J-Dog
May 7, 2009

Mark – Thank you for doing the dirty, nasty but necessary work of de-coding a Dembski diatribe. It would be great if holding him up to riducule for his fashion sweater choices would shut him up, but since he has no shame, your solid thrashing of his scholarship (or lack of it) is much appreciated.

9. #9 sng
May 7, 2009

Somewhat offtopic question but, with any luck, not all that far offtopic.

Are there any resources, online or in ebook format by choice, for a not very mathematical but not unbright nerd to start learning the basics of information theory?

10. #10 Bob O'H
May 7, 2009

which I take both to be a sign of lack of professionalism,

Marks does have an immature sense of humour, so this is no real surprise. I think they’re also playing games after olegt, in a rather good bit of design detection involving pdfs, found out that Marks was “Galapagos Finch”.

I view this side of things as a game, so I found it quite funny. which probably says a lot about my sense of humour too.

11. #11 Dave Wisker
May 7, 2009

Evolution is a dynamic landscape – that is, a landscape that changes in response to the progress of the search. Pretty much every argument that Dembski makes can be thrown out on the basis of this one fact: all of his arguments are based on static landscapes. Once the landscape can change, every single one of his arguments become invalid – none of them work in dynamic landscapes.

I prefer the seascape metaphor myself. Peaks rise,fall, combine into bridges between other peaks, and create deep valleys in a constantly changing way.

12. #12 Left_Wing_Fox
May 7, 2009

If you delayer the PDF to see the text under the blackouts, it says “International Journal of Fun and Games – You’re too Clever – Gotcha!” – which I take both to be a sign of lack of professionalism, and an admission that they can’t get this stuff published anyplace real.

That is the silliest thing I ever saw. It’s like a 5-year old playing ‘Pick a hand’, then squirming desperately to shift the candy behind their back when you choose the one holding the candy.

13. “That is the silliest thing I ever saw. It’s like a 5-year old playing ‘Pick a hand’, then squirming desperately to shift the candy behind their back when you choose the one holding the candy.”

Welcome to Intelligent Design!

14. #14 Richard Simons
May 7, 2009

[Dawkins’ weasel is] a tiny, silly, throwaway example that demonstrates one minor feature of how evolutionary processes work; it was never intended to be a complete example of how evolution produces life.

It might be a trivial demonstration to you, but sorting it out kept the people at Uncommon Descent occupied for weeks. 🙂

15. #15 MPL
May 7, 2009

Re: No. 1.

You’re right. In fact, to go even further, the fact that evolution proceeds by mutations of existing organisms implies that evolutionary histories (i.e. “search paths” if you have it Dembski’s way) will show near-continuity. Since, in fact, this is exactly what we do see, it would seem that the NFL argument actually strongly supports evolution.

16. #16 Thomas S. Howard
May 7, 2009

@14

Yeah, and after those weeks of effort they still don’t understand how Weasel actually works or that they’re painfully wrong. They never really sorted anything out so much as they all just gave up after a bit of mutual reassurance that they’d had it right all along.

17. #17 isotelesis
May 7, 2009

Rather than strictly teleonomic or teleological evolution, I would refer to the process as fundamentally more subtle than an “all-knowing designer”, and a “program executing automaton”, evolution/natural selection is here to stay, but those who try applying it to explain everything are not true intellectuals. The idea that somehow our seed was planted here by some intelligent aliens isn’t that absurd, but the question becomes what created them? Eventually you get down to patterns, which sometimes have their own natural ‘design’ that morphs with the help of randomness and other factors…the question is whether the design was pre-planned or self-assembling…in some cases it could have been either, there is no evidence for an exact mechanism for reflexive design, that doesn’t mean in principle there couldn’t be features which permit degrees (crude to refined) of self-actualized autopoietic innovation. I think Debemski is not the person you want to cite if you believe Darwinism was unsuccessful at eradicating the need for Aristotle’s ‘final’ cause, Daniel M. Dubois is a non-ID scientist who nonetheless has some excellent work on so-called morphotelic ‘phenomena’.

Towards an anticipatory view of design:

“Rather than confining itself to theological or teleological causation, ID theory technically allows for any kind of intelligent designer – a human being, an artificial intelligence, even sentient aliens. This reflects the idea that intelligence is a generic quality which leaves a signature identifiable by techniques already heavily employed in such fields as cryptography, anthropology, forensics and computer science. It remains only to note that while explaining the inherent complexity of such a material designer would launch an explanatory regress that could end only with some sort of Prime Mover, thus coming down to something very much like teleology after all, ID theory has thus far committed itself only to design inference. That is, it currently proposes only to explain complex biological phenomena in terms of design, not to explain the designer itself. With regard to deeper levels of explanation, the field remains open.”

18. #18 RBH
May 7, 2009

Mark said, “Search is a lousy model for evolution; general search is a particularly lousy model.”

Thank you, thank you, thank you! I’ve been saying that the search metaphor for evolution is a snare and a deception for years.

And there’s another reason the search metaphor is deceptive: Biological evolution doesn’t test randomly over all possible nodes, but rather preferentially explores (via mutations) nodes that are in the neighborhood of an already successfully reproducing population. One can give that more rigor, e.g., by defining “neighborhood” in terms of all nodes accessible by single applications of the various mutation operators. But it’s intuitively clear without that.

19. #19 B L Harville
May 8, 2009

Did Dembski and Marks manage to publish this in a real journal?

No. This is separate from the other two “papers”. Believe it or not, this “paper” is going to be a chapter in one of Dembski’s upcoming books. A book chapter written in the form of a scientific paper. Who else who write something like this? And why would Dr. Marks, who apparently has a real job in academia at a real university, be involved with this character?

20. #20 Gunthernacus
May 8, 2009

Which comes back to one of the whole objections to the whole NFL-based approach: Dembski insists on searches that perform equally well over all possible landscapes. Saying that a search algorithm performs only on some set of search spaces is exactly the same as saying that the search algorithm contains information about those search spaces!

Thanks for that.

21. #21 Tyler DiPietro
May 8, 2009

RBH,

“And there’s another reason the search metaphor is deceptive: Biological evolution doesn’t test randomly over all possible nodes, but rather preferentially explores (via mutations) nodes that are in the neighborhood of an already successfully reproducing population.”

I’m not a huge fan of the search metaphor either, but what you’re describing here sounds like your standard issue hill-climber, a very common form of local search.

As to Dembski’s paper, his whole concept of “active information” is circular. He claims that active information is needed to increase the probability of success in a search, but he defines active information with the quotient of the probabilities of success of a “null search” (Dembski’s pretentious term for a random walk) and an “alternative” search that is defined to be more successful. The LCI, which is basically the notion that one has to increase this “active information” to increase the probability of success in a search, reduces to mere tautology. Not exactly “meaningless”, but definitely. And obviously irrelevant to biological evolution.

22. #22 Tyler DiPietro
May 8, 2009

“Not exactly “meaningless”, but definitely [trivial]. And obviously irrelevant to biological evolution.”

Fixed.

23. #23 isotelesis
May 8, 2009

A non deterministic hyperincursive field rises in a self-reference Fractal Machine. This can be related to the Final Cause of Aristotle.

http://www.evol.nw.ru/labs/lab38/levchenko/articles/belgy2fin.htm

24. #24 Joe G
May 8, 2009

Instead of complaining about the Dembski/ Marks paper why don’t you focus on supporting YOUR position?

IOW if you could support the claim that nature, operating freely can produce what Dembski/ Marks says only agencies can produce, then they are refuted by the DATA/ reality.

So the bottom line is all you have to do is start supporting the claims of YOUR position and Dembski/ Marks won’t have anything to write about.

But I take you can’t and so here we are…

25. #25 Mark C. Chu-Carroll
May 8, 2009

Re #23:

You seem to be rather confused about the background to all of this.

This paper is part of Dembski’s attempt to argue away all of the data that’s been provided about what evolutionary processes can produce. Real scientists (ie, not me – I’m just a software engineer) have done lots of real, fascinating work in a large number of fields to help understand evolution. The resulting evidence ranges from computational simulation, to genetic analysis, to observed evolution, to ERV relationships, to lord only knows what else.

This work by Dembski and Marks is an attempt to argue that all of that evidence – every simulation, every observed event, every genetic relationship, every gene sequence, every mathematical proof – that all of it is wrong.

You see, the actual scientists have accumulated, presented, written about, and studied virtual mountains of evidence. Dembski and his DI friends have *never* done a shred of science; have never collected a single shred of evidence. There’s not one page of actual data that’s ever been produced by Dembski &co. Not one actual scientific study of anything.

The reason that Dembski and Marks have something to write about – despite the fact that they’ve done *zero* research studies, zero experiments, and zero data collections – is because the actual scientists have done the work.

And as you can see in my writeup above, their critiques of the work of actual scientists are pathetic.

26. #26 Evil Bender
May 8, 2009

Scientists: look at all these examples we’ve found of evolution working. Here, for example, are some some transitional forms, and here are laboratory experiments, and here are examples of evolution in action from nature. Cool, huh?

IDiots: You have not provided enough detail. MOAR “pathetic” detail is required! ID is true. We’re not interested in doing the research to prove that, by the way.

Scientists: *Too busy doing science to respond*

IDiots: still waiting for that evidence.

Scientists: Here’s a lot more cool evidence we’ve found about evolution.

Joe G: I can’t hear you! I’m too busy moving the goal posts!

27. #27 abb3w
May 8, 2009

Tyler DiPietro: I’m not a huge fan of the search metaphor either, but what you’re describing here sounds like your standard issue hill-climber, a very common form of local search.

I think a “valley-descender” might be better in this context, as on Earth (on land) it’s valleys that are usually more conducive to life.

Also, evolution is a random walk… but a non-uniform probability random walk, in an n-dimensional space. I find it’s easiest to visualize using one dimension of “fitness” (available exergy), one of local population, two of genome space, two of physical space, plus one of time. Life falls down into local valleys, fills them up, and spreads out from them, tending to make its way along as deep a shallow “river” as it can find until it reaches a new (and hopefully “deeper”) valley.

Essentially, any information that accumulates is via signal “transmitted” from the landscape; eg, the circa 11.6 μHz light-and-dark signal of Earth’s spin.

28. #28 B L Harville
May 8, 2009

IOW if you could support the claim that nature, operating freely can produce what Dembski/ Marks says only agencies can produce, then they are refuted by the DATA/ reality.

Creationists are constantly claiming that there is no evidence for what they call “macro-evolution”. There certainly is – you just don’t accept it. That’s your problem, not the problem of the scientific community. The fossil record clearly shows that “macro-evolution” has taken place. And if you are going to claim that the fossil record is the result of intelligent designers or evil demons it’s your job to show that such things actually exist. Good luck.

29. #29 trrll
May 8, 2009

I actually rather like the “hills and valleys search” metaphor, but it is misleading in a couple of ways. One which Mark has written about is that the hills and valleys are not static, as they are in part comprised of interactions with other populations that evolve and coevolve.

Another misleading aspect is that the metaphor tempts one to think of the hills and valleys as independent of the organism, but a major, perhaps dominant, factor is the coupling between genotype and phenotype, which influences the shape and “smoothness” of the phylogenetic landscape. That is, organisms with developmental and regulatory feedbacks capable of compensating for changes in the function of individual genes will exhibit a “smoother” phylogenetic landscape, in which steepest-descent/ascent search algorithms will be most successful. And of course, modern organisms are the descendants of those that were most successful in evolving. So there is selection for the capacity to evolve by natural selection.

So evolution not merely optimizes the organism with respect to the “landscape,” but actually optimizes the landscape itself toward ones in which evolutionary search algorithms are most effective.

30. #30 Darby M'Graw
May 8, 2009

A book chapter written in the form of a scientific paper. Who else who write something like this?

This reminds me of Jesus is like my Scanning Electron Microscope by Mark Armitage. Although this is a personal testimonial with only the barest resemblance to a science book, the Acknowledgements page reads: “The author greatly acknowledges the assistance of the anonymous reviewers of this manuscript.” That was a real hoot. Almost as much fun as the disclaimer on the copyright page: “This is a work of fiction. All of the characters, names, incidents, organizations, and dialogue in this novel are either the products of the author’s imagination or are used fictitiously.” If only all works of religious apologetics were as honest.

EXTRA! EXTRA! Some movement in the Amazon reviews! My formerly lone review has now been joined by another, a five star review – from the book’s author!

31. #31 RBH
May 8, 2009

Tyler DiPietro wrote

I’m not a huge fan of the search metaphor either, but what you’re describing here sounds like your standard issue hill-climber, a very common form of local search.

Sorry, I boiled it down too much and wasn’t clear. The point of my comment is that the denominator of any probability estimate associated with sampling some node is not uniform, but rather is biased to preferentially sample the neighborhood of a population already known to be viable. That is, the population is already at a ‘goal,’ a ‘solution,’ a local optimum. It starts at a viable (cluster of) node(s), and doesn’t move unless drift (see below) or alterations in the selective environment (changes in the topography of the landscape) create new ‘opportunities’ for the population. The probability of finding another viable node when the landscape is deforming is greater in that neighborhood than on some random point in the whole space. As a consequence, any probability estimator that has as its denominator the number of nodes in the whole space is bullshit, since the whole space isn’t sampled. That’s related to, but distinct from, the local hill-climbing search issue.

Further, in high-dimensioned spaces, as Gavrilets has shown, the hill-climbing notion itself is suspect, since the space is shot through with ridges rather than the typical image of hills and depressions.

32. #32 rimpal
May 8, 2009

What the heck is intelligence? While Bill waffles and spins about obfuscating Shannon, Kolmogorov etc., about information, he is yet to come up with any operational definition of intelligence.

33. #33 isotelesis
May 8, 2009

‘What the heck is intelligence?’

Dembski’s work is tainted by ideology, which is why I wouldn’t take him seriously, the interesting work is being done in cybernetic explanations of emergence.

http://www.emeraldinsight.com/Insight/viewContentItem.do?contentType=Article&contentId=1640461

“ID theory technically allows for any kind of intelligent designer – a human being, an artificial intelligence, even sentient aliens. This reflects the idea that intelligence is a generic quality which leaves a signature identifiable by techniques already heavily employed in such fields as cryptography, anthropology, forensics and computer science.” -Chris Langan

34. #34 dean
May 8, 2009

The Chris Langan who said this about creationism?

“”I believe in the theory of evolution, but I believe as well in the allegorical truth of creation theory. In other words, I believe that evolution, including the principle of natural selection, is one of the tools used by God to create mankind. Mankind is then a participant in the creation of the universe itself, so that we have a closed loop. I believe that there is a level on which science and religious metaphor are mutually compatible”

Or who said this?

“People who wanted to have children would apply to make sure they have no diseases. Why do we have to do it through genetic engineering? Well, we have to let only the fit breed…. Freedom is not necessarily a right. It is a privilege that you have to earn. A lot of people abuse their freedom and that is something that people have to be trained not to do.”

When asked who would do the training, he said

“Well, I’d be perfectly willing to do it myself. Just put me in charge.”

That Chris Langan – former bouncer? How is he any better than the typically ignorant creationist?

35. #35 G.D.
May 8, 2009

A (not so) minor nitpick. You say:

“You can derive really incredibly looking equations using perfectly valid proofs, to demonstrate any conclusion that you want. But your conclusion is only valid in a setting where you’ve accurately modeled the reality that you’re trying to describe.”

Nope. First, and least important, it is very strange and close to a category mistake to say that a conclusion is “valid”. Second, and much more important, the validity of an argument is independent of the truth or falsity of the premises. I think what you mean is “sound”, or something similar (i.e. a valid argument with true premises and (hence) a true conclusion) – or are you here just reverting to a colloquial use of “valid” (I kinda think so, since this is logic 101, 1st meeting).

36. #36 isotelesis
May 9, 2009

Chris Langan is an odd character, obviously IQ doesn’t necessarily measure real-world functioning, I was simply quoting a clear definition of what ID theory technically is.

Are you aware that creationism is a different than ID-theory? You seem to be mixing the two up, there are creationists who also believe in the theory, and there are non-creationists who believe there is more to Aristotle’s final cause than an Omega Point, and has more to do with a phenomenon known as ‘Hyperincursion’. There is no final cause, but there is something very similar, go on and make fun of Chris Langan all you want, that is beside the point, he is arrogant and egotistical if you ask me, and has a little too much faith in his limited knowledge of genetics.

“Darwin’s theory of natural selection made any invocation of teleology unnecessary. From the Greeks onward, there existed a universal belief in the existence of a teleological force in the world that led to ever greater perfection. This “final cause” was one of the causes specified by Aristotle. After Kant, in the Critique of Judgment, had unsuccessfully attempted to describe biological phenomena with the help of a physicalist Newtonian explanation, he then invoked teleological forces. Even after 1859, teleological explanations (orthogenesis) continued to be quite popular in evolutionary biology. The acceptance of the Scala Naturae and the explanations of natural theology were other manifestations of the popularity of teleology. Darwinism swept such considerations away.” -Ernst Mayr

37. #37 RBH
May 9, 2009

G.D. wrote

Nope. First, and least important, it is very strange and close to a category mistake to say that a conclusion is “valid”. Second, and much more important, the validity of an argument is independent of the truth or falsity of the premises. I think what you mean is “sound”, or something similar (i.e. a valid argument with true premises and (hence) a true conclusion) – or are you here just reverting to a colloquial use of “valid” (I kinda think so, since this is logic 101, 1st meeting).

Well, I won’t defend the terminology mentioned, but the point seems clear: to the extent that the terms and operators of a math model don’t map veridically to the objects, properties, and processes identified by a substantive theory as relevant, a perfectly valid (in the sense of syntactically correct) set of manipulations of those terms and operators can yield nonsense when the result of the manipulations is mapped back to the real-world system allegedly being modeled.

38. #38 trrll
May 9, 2009

Are you aware that creationism is a different than ID-theory? You seem to be mixing the two up

While the Discovery Institute has been working very hard to try to dissociate ID from Creationism, Barbara Forrest proved beyond a shadow of a doubt that “Intelligent Design” is just a rebranding of Creationism. Her evidence played a major role in convincing the Bush-appointed judge in the Dover trial that Intelligent Design was a religious doctrine rather than a scientific theory. Indeed, she was able to prove that the defining Intelligent Design text Of Panda’s and People was originally written as a Creationist text, and became a text of “Intelligent Design” by search/replace when the Discovery Institute decided on the rebranding.

The Discovery Institute now would like to rewrite history to pretend that the term “Creationism” refers only to what used to be called “Young Earth Creationism,” to distinguish it from “Scientific Creationism” (the form of Creationism that is now renamed Intelligent Design). Of course, nobody who had followed the debate for any period of time was fooled by this, because we all remember that every single argument now advanced in favor of “Intelligent Design” was originally articulated by people who characterized themselves as “Creationists.”

39. #39 isotelesis
May 9, 2009

So ID-theory is bogus, I don’t really care, anyone care to comment on the ideas of Robert Rosen?

I’ll post the link again.

40. #40 isotelesis
May 9, 2009

To avoid the obfuscative nature of ‘intelligent design’, perhaps we should call the type of functions associated with hyperincursion as simply ‘reflexive design’, since most fail, but to some degree the organism incorporates some kind of temporal logic when it comes to responses of a cybernetic nature. You can’t avoid the design factor, some a poorly developed, others more keenly. This wouldn’t contradict Darwinian thinking, just extends it to reflect what scientists have learned about complex adaptive systems, as Murray Gell-Mann liked to call it: ‘plectics’, thus allowing us to consider enfolded quantum-holographic multiphasic oscillation…go ahead and label me a crackpot, there are too many good ideas out there for me to care what you think.

41. #41 isotelesis
May 9, 2009

Perhaps the only thing ‘RD’-theories require is a system of biosemiosis and self-reference.

42. #42 trrll
May 9, 2009

So ID-theory is bogus, I don’t really care, anyone care to comment on the ideas of Robert Rosen?

Aside from the use of the word “design,” it seems pretty much off-topic.

At first glance, it looks like an example of taking a concept that is not particularly novel and larding it up with a bunch of formalism that doesn’t add any additional insight.

43. #43 isotelesis
May 9, 2009

Design is the outcome pattern, it can be planned by various means and degrees or not, it can be accidental, is can be a product of organization, autocatalyzing reactions, a random program, or even holographic, reflexive, and fractal in quality.

Rosen’s work has to do with organisms taking into account predicted future states in the configuration/modulation of present ones, this is very similar to what teleology describes. It does not mean the two concepts are uniquely isomorphic. This is more about philosophy of theories and models, Darwin was a natural scientist, Aristotle and Kant were mostly philosophers, Norbert Weiner was better than all of them, basically he was a mathematician, and his theory of cybernetics has applications to evolutionary models. ID theory invokes intelligence as the prime mover, RD refers to primary movements as hyperincursions, which demonstrate morphotelic properties (teleonomic and in some cases teleologic.) Darwinism only explains teleonomy, cybernetics/plectics seek to explain ‘teleology’, and even more interested behaviors in complex adaptive systems consisting of many agents…thus it becomes game-theoretic and even polytelic. I hope you won’t call it lard, and try to understand what I’m talking about, for the sake of science.

‘Teleonomy is the quality of apparent purposefulness and of goal-directedness of structures and functions in living organisms that derive from their evolutionary history and adaptation for reproductive success.
The term was coined to stand in contrast with teleology, which applies to ends that are planned by an agent which can internally model/imagine various alternative futures, which enables intention, purpose and foresight.’

44. #44 Blake Stacey
May 9, 2009

This reflects the idea that intelligence is a generic quality which leaves a signature identifiable by techniques already heavily employed in such fields as cryptography, anthropology, forensics and computer science.

Um, no. A perfectly encrypted message — for example, one enciphered using a one-time pad — is indistinguishable from noise. Anthropology and forensics rely on much more than the presumption of intelligence: they generate hypotheses based on the premise that the intelligences at work were human, with all the baggage that implies.

45. #45 isotelesis
May 9, 2009

‘A perfectly encrypted message — for example, one enciphered using a one-time pad — is indistinguishable from noise.’

That could imply that even if there were some vague notion of something like ‘intelligence’ participating in the fabric of unfolding events, it would be indistinguishable from noise anyway?

46. #46 trrll
May 9, 2009

‘Teleonomy is the quality of apparent purposefulness and of goal-directedness of structures and functions in living organisms that derive from their evolutionary history and adaptation for reproductive success.
The term was coined to stand in contrast with teleology, which applies to ends that are planned by an agent which can internally model/imagine various alternative futures, which enables intention, purpose and foresight.’

Yes, I just can’t get particularly excited about constructing a formalized taxonomy of concepts that are not particularly novel. Now if this sort of formalism were to yield some sort of real insights that can be validated by programming improved artificial intelligence systems, or by making novel predictions regarding neuronal connectivity or synaptic dynamics in the brain, then I’d start to take interest.

47. #47 isotelesis
May 9, 2009

Indeed cognitive computation is literally where the action is.

I hate religion btw, so I’m not excited about seeing future scientists learning from parochial folk, they deserve better.

Basically the only way I can technically see the I’D argument working is by clarifying definitions and semantics, I am not interested in taxonomies.

The least fluffy version of I’D theory would amount to: intelligence exists in the universe, it played a role in evolution, whether than was hunting/gathering strategy or some kind of biosemiotic process involving neomorphic mutations, or something even more bizzare like ‘smart’ energetic clouds seeding or altering the course of life on planets and not caring what happens to them afterwards, the field is almost too open to interpretation. Which is why I prefer describing the cosmos as reflexive, not necessarily intelligent, however we nonetheless play some active role in evolution, where as a result of intelligence, selfishness, or lack of intelligence (Darwin Awards.)

48. #48 Michael Ralston
May 9, 2009

isotelesis: If ID was parsed to mean “intelligence has affected evolution”, it would be a trivial statement (because of the last several thousand years), but would be entirely uninteresting.

And it wouldn’t support any of Dembski’s or Behe’s claims.

And it wouldn’t damage evolutionary theory at all, since evolutionary theory already accounts for the known parts of that effect, and could in principle account for it in almost any form short of “God is responsible for everything” – which renders ID moot and useless as a theory.

But much of what you’re saying seems to consist of throwing together buzzwords without regards to their actual meanings.

49. #49 isotelesis
May 9, 2009

‘since evolutionary theory already accounts for the known parts of that effect, and could in principle account for it in almost any form short of “God is responsible for everything”‘

So evolution via natural selection in principle explains the emergence of any form of intelligence capable of engineering another life form for entertainment? Intelligent design is not competing with Darwinian natural selection. It has simply expanded on alternative possibilities. Since there is no way to infer whether our ancestral species were intentionally set up to make it up to this point, the only viable explanation is a multiversal landscape as described by Susskind.

Which ‘buzzword’ was misappropriate?

I am simply discussing the reasons why Dembski and his colleagues have found the preemptive exclusion of ‘teleological’ causes in evolution as unjustified. I used ‘hyperincursion’ as an alternative. They have different motivations, some religious, others philosophical. Rosen discusses how modeling biology in terms of ‘failure modes’ doesn’t satisfactorily explain important aspects of biological evolution. I believe what I am defending is a biosemiotic and cybernetic view of evolution which considers the process of adaptation in complex systems as more than a measure of survival.

50. #50 isotelesis
May 9, 2009

While I do not agree with Dembski’s approach, I recognize that evolutionary biology is expanding, I hope some of the buzzwords I mentioned were new to some of you.

http://en.wikipedia.org/wiki/Biosemiotics

51. #51 trrll
May 10, 2009

So evolution via natural selection in principle explains the emergence of any form of intelligence capable of engineering another life form for entertainment?

Obviously, considering that we are in fact doing this ourselves. In general, biologists are not necessarily hostile to the notion of life on earth having been engineered by some extraterrestrial form of life. But in the absence of compelling evidence, it just seems like an unnecessary complication, merely displacing the interesting problem to elsewhere, since science would then need to explain where that form of life came from and how it came to be. Occam’s Razor teaches us not to add complications to our hypotheses unless and until they are compelled by the data.

I am simply discussing the reasons why Dembski and his colleagues have found the preemptive exclusion of ‘teleological’ causes in evolution as unjustified.

Nothing has been “preemptively excluded” by biology. Evolution on earth is simply favored as being the simplest hypothesis consistent with the data. Dembski and other ID/Creationists are rather making the opposite argument, that evolution can be preemptively excluded, and that life must therefore have been intentionally engineered. The real problem with ID is that it stops there. It makes no attempt to construct testable hypotheses as to the origin or nature of the supposed “Designer.” Scientifically, this makes no sense, of course. One does not arbitrarily draw a conceptional boundary and declare “beyond this point thou shall not go,” because this creates a conceptual dead end that is an obstacle to scientific discovery. But as demonstrated by Barbara Forrest, ID is not a scientific theory, it is a legal dodge–they explicitly exclude discussion of the nature of the Designer because to do so would expose ID as a stalking horse for the religious doctrine of Christian Creationism.

52. #52 Hermagoras
May 10, 2009

Mark, Dembski has responded (sort of) to this post. (His response forced me to read his paper — my god what a windbag.) I hope you’ll respond to his response.

53. #53 trrll
May 10, 2009

While I do not agree with Dembski’s approach, I recognize that evolutionary biology is expanding, I hope some of the buzzwords I mentioned were new to some of you.

Biology has no objection to jargon, as reading any paper on molecular genetics or intracellular signal transduction will quickly convince you. But scientists tend to adhere to a kind of Occam’s Razor here as well–that is, they are reluctant to accept new jargon unless there is some kind of immediate payoff in terms of experimentally testable hypotheses. So if you can come up with some sort of important experimental result, and can convince the reviewers of your publication that your hypothesis or results can be most clearly and economically described using your new terminology or conceptual framework, then scientists will enthusiastically embrace it.

54. #54 Thomas Groover
May 10, 2009

Mark: could you possibly make your reviewing style just a little more foul? Also I might really benefit from believing Dembsky and Marks are as stupid as you say, and you’re personal style isn’t quite foul enough to get this across. I mean you could give examples of stupid people who’ve mastered higher mathematics but who who can’t graduate from high school proving they’re stupid. That way I could believe these guys are really stupid.

Oh– and it might also help if you, when writing serious scientific reviews, would use terms that go beyond those such as excreta and possibly give favor to those such as excreta buzzing with flies having an affinity for (fill-in-the-blank researcher’s name)’s face. This way even more people will take you seriously than do now.

Thomas Groover, MSEE (UT Austin ’81)
Houston, TX

55. #55 LimboAZ
May 10, 2009

I have a question for Mark Chu-Carroll. Can you honestly tell us that you read the entire paper carefully and made an attempt to understand it fully? After reading both the paper, and your response, I find that very hard to believe. Your response is mostly a straw man rant filled with ad hominem insults like “intellectual excreta”, “same old rubbish”, “not real scientists”.

Quite frankly, when you do get specific with criticisms, they don’t accurately reflect what the paper is actually arguing.

The weakest aspect of the TOE is how random mutations create complex information. Everybody understands how selection preserves and expands useful genetic information. I, like you, am a software engineer. We of all people should understand how bad randomness is at creating complex, useful and meaningful code.

I, for one, would appreciate it if you could actually deal with the facts as presented. You lose credibility with all the insults. From my experience, it is usually the person losing the argument that resorts to such tactics.

56. #56 dean
May 10, 2009

“Are you aware that creationism is a different than ID-theory”

Are you aware that is a complete lie?

57. #57 dean
May 10, 2009

” Intelligent design is not competing with Darwinian natural selection.”

No, intelligent design is being used by people who don’t like, or understand, science, as a tool to wedge their own ideas about religion into classrooms.

Intelligent design is simply a poor disguise put over the idea of creationism – and no articles full of loopy prose can change that fact.

58. #58 trrll
May 10, 2009

The weakest aspect of the TOE is how random mutations create complex information. Everybody understands how selection preserves and expands useful genetic information. I, like you, am a software engineer. We of all people should understand how bad randomness is at creating complex, useful and meaningful code.

Of course, biologists know that randomization/selection is perhaps the most powerful mechanism for creating novel information==perhaps even the only mechanism. Biologists routinely use randomization/selection methods to obtain information that the smartest people in the world have been unable to figure out. Indeed, the most ironic thing about the ID contention that intelligence is more powerful than randomization/selection is that the neuroscience strongly suggests that human creative intelligence is itself a manifestation of randomization/selection mechanisms that are analogous in many ways to evolution.

59. #59 Bayesian Bouffant, FCD
May 10, 2009

LimboAZ: I, like you, am a software engineer. We of all people should understand how bad randomness is at creating complex, useful and meaningful code.

You must be an incompetent software engineer if you are not familiar with the fabulous successes achieved through the use of genetic algorithms.

60. #60 Herod the Freemason
May 10, 2009

isotelesis: Intelligent design is not competing with Darwinian natural selection.

Not in the peer-reviewed literature, anyway. They’d rather go straight to Fundamentalist-controlled school boards than do any actual research.

61. #61 isotelesis
May 10, 2009

“Of course, biologists know that randomization/selection is perhaps the most powerful mechanism for creating novel information==perhaps even the only mechanism.”

The constraints on indeterminacy, unbounded choice, unbounded nondeterminism, hypercomputing, etc. would be interesting subjects in the studying of biological systems. There are so many better approaches to studying life than throwing one’s hands up in the face of complexity and depending on a supernatural creator.

Cause and Effect in Biology: http://faculty.washington.edu/lynnhank/Mayr2.pdf

System attributes like complexity, connectivity, intransparency, dynamics, and polytely exhibit patterns was first described by the cognitive scientist Dietrich Dörner:

62. #62 isotelesis
May 10, 2009

There may be no such thing as “intelligent design”, but a study of “intelligent designs” would be helpful for any evolving organization.

63. #63 isotelesis
May 11, 2009

For the dispassionate intellectuals:

We should distinguish between design by(operator), with(reflexivity), or producing(generated) intelligence, and how signification processes (de-sign) and instruction methods(in-tel’ligence) are relevant in terms of universal (by), holoversal (with), and multiversal (producing) unfoldings; And where is the line we don’t cross, as in the range of modelable features or technical assumptions.

I can’t really swallow what Dembski has written, because it takes some of these considerations for granted, we also should clarify which sort of global distinctions are moot for the purposes of simulated realities.

64. #64 isotelesis
May 11, 2009

For the concerned parents:

Government funded schools have a responsibility to uphold a certain standard, one size does not fit all, therefore private schools/home schools are a choice. Lets say your kid has special abilities, and they need something explained to them either more simply/metaphorically/allegorically with stories, as a parent it is your perogative to tell them about the birds and the bees while public school kids go look it up on the evil internet. Or lets say they are like William J. Sidis and can’t be contained by the boundaries imposed on them in public schools, well thanks to the holy internet they can be enriched to their hearts content.

65. #65 dean
May 11, 2009

“well thanks to the holy internet they can be enriched to their hearts content.”

You’re really just yanking chains now, aren’t you – you are too consistently odd to be serious.

66. #66 Jeff Alexander
May 11, 2009

Dembski claims:

Exhaustively or blindly searching a space this size to find a target this small is utterly beyond not only present computational capacities but also the computational capacities of the universe as we know it.

In my last place of employment we routinely examined high dimensional search spaces that had more than 10300 possibilities. Of course we used genetic algorithms to do it which is neither exhaustive nor blind.

67. #67 LimboAZ
May 11, 2009

Bayesian Bouffant: You must be an incompetent software engineer if you are not familiar with the fabulous successes achieved through the use of genetic algorithms.

GA’s are a useful technique but require knowledge to create fitness functions that can test intermediate results, otherwise the search becomes unwieldy fast. GA’s are basically a random search with the aid of heuristics and are therefore still subject to the laws that govern probabilities. This is the whole crux of the problem when dealing with random searches. Are you arguing otherwise? That would be amusing.

68. #68 Mark C. Chu-Carroll
May 11, 2009

Re #66:

You can define GA search as “random search with heuristics” if you want, but it’s not a particularly good definition – it’s basically just a way of defining the search to try to minimize the role of selection – because the selection process (aka the fitness function) is what really does the work.

So why call it a heuristic rather than a selection function? I can’t see any particularly good reason. It obscures rather than clarifies – because most people think of heuristic search in the AI sense – where it basically means applying the heuristic to select search
paths; whereas in GAs, the selection function works *after* a search step to select the most successful paths. They’re equivalent in a theoretical sense, but they’re typically fairly different in practice.

I can only see two reasons for calling GA as heuristic search:
(1) You’re doing formal studies of search systems (in which case you’d
categorize AI heuristic search and GA as being pre- and post- variants of
the same method; or
(2) You want to try to obscure the fact that you’re using selection.

When you describe GA as being really just random search with heuristics, I strongly suspect that you’re trying to pull the latter – because “random search with heuristics” sounds more like something that requires intelligence than “genetic algorithms”.

69. #69 isotelesis
May 11, 2009

“you are too consistently odd to be serious.”

I’m just making fun of the whole stupid debate over what should be taught in school, I compare it to a parent who prefers to explain that reproduction is a wonderous act which involves higher powers that ‘guide’ to some degree the course of ‘evolution’ and one prefers a child have a realistic and less metaphysical interpretation, aka a bunch of bald monkeys had sex.

70. #70 mechape
May 11, 2009

In the paper Dumbsky and Marks wrote :”The LCI regress suggests a deep connection between the Law of Conservation of Information and the Second Law of Thermodynamics. ”

I can’t verify statistical-mathematical rigournes of their LCI theorems, but it seems that even if it is logically correct from mathematics than ID is simply our SUN. That is the source of “additional information”. Frankly, it is quite an old belief – Sol was deified long before Jesus 🙂 Other possibility that all their mathematical apparatus doesn’t hold scrutiny, than it is an utter crap.

71. #71 isotelesis
May 11, 2009

‘higher powers’ implies everything we haven’t understood.

72. #72 Tyler DiPietro
May 11, 2009

“GA’s are basically a random search with the aid of heuristics and are therefore still subject to the laws that govern probabilities.”

“Random search with the aid of heuristics” is literally gibberish. Try again.

73. #73 Mark C. Chu-Carroll
May 11, 2009

Re #71:

Not so fast, Tyler. Random search with heuristics isn’t gibberish. In fact, it’s a fairly common algorithm in AI.

The idea is that you have a search landscape defined by a successor relation – so from any state in the search, there is some set of successors. You want to find a path to an optimum value.

If the successor relation is sufficiently large – meaning that each state has a very large number of possible successors, and the
landscape is sufficiently densely connected (meaning that you’ll
likely eventually hit the optimum if you keep searching and avoiding states that you’ve already visited) – then exploring all of the successor states is too expensive. So you use a semi-random strategy. You select a sample of the possible successors, and evaluate them, and follow through on the ones that look best.

How do you select the successors to explore? You can use pure randomness, or you can use a heuristic. If you use a heuristic, it’s still basically a random search – but you use a heuristic to try to speed it up. You’re still bounded by the performance of a random search, but hopefully the heuristic will help.

74. #74 Tyler DiPietro
May 11, 2009

My bad, Mark. My familiarity with the use of the term “heuristic” in CS is a search algorithm for which precise mathematical performance properties are not sufficiently pinned down (e.g., tabu search, simulated annealing and, most pertinently, evolutionary computation).

75. #75 Tyler DiPietro
May 11, 2009

One second thought, I think all such search heuristics are random in the sense that a successive state in the algorithm isn’t entirely determined by it’s predecessor. I think my mistake was conflating “random” with “uniform probability distribution.”

76. #76 Drosera
May 12, 2009

Dembski’s protein argument is just another version of the stupid ‘tornado producing a Boeing 747 out of a scrap yard’ metaphor. No organism is ever faced with the challenge to assemble a completely new kind of protein consisting of 100 or so amino acids from scratch. Complex proteins have evolved from simpler proteins (and also vice versa), just like complex organs have evolved from simpler ones, through the same mechanism of mutations, gene duplication, etc., followed by selection.

Only complete idiots take ID seriously. Dembski himself is a waste of proteins.

77. #77 fnxtr
May 12, 2009

Have you read Iain Bank’s The Algebraist? The aliens do something with their algebra very much like what it looks like Dembski has done with his whatever-it-is.

78. #78 Ryan Cunningham
May 13, 2009

Mark, do you know if anyone has ever tried characterizing the landscapes a realistic model of evolution would perform well on? Even if search algorithms aren’t the best system for studying evolution, there’s certainly a hint of optimization in natural evolution. I recall reading lots of genetic algorithm research papers trying and failing to do this for their very unnatural model of evolution. It seems like even very crude characterizations of these landscapes would be extremely useful for understanding evolution, but I don’t know of any successful attempts to explore this idea.

79. #79 Erik 12345
May 14, 2009

I’m a bit concerned that Chu-Carroll’s piece contains some misuse of the NFL theorems too. E.g. the connection claimed here between (finite?) representations of real numbers and the NFL theorem is surely not correct:

Unfortunately, most real numbers are undescribable. There is no notation that accurately represents them. The numbers that we can represent in any notation are a miniscule subset of the set of all real numbers. In fact, you can prove this using NFL.

All real numbers can be represented as limits of rational numbers (or as infinite decimal expansions, if you wish). That not all real numbers can be represented using, say, finite decimal expansions has more to do with “size” (cardinality) of the set of real numbers than the NFL theorem.

80. #80 Stephen Wells
May 15, 2009

@78: I don’t think Mark said that you can’t prove the point _without_ NFL: only that you can prove it _with_ NFL.

81. #81 slpage
May 15, 2009

So, essentially, Dembski is saying that a tree really does NOT make a sound when it falls if no human is there to hear it….

82. #82 snaxalotl
May 16, 2009

oh good. I’ve been waiting for someone to make a comparison with real numbers. a very similar way of putting it is to say that most real numbers, like most “possible” landscapes, are incompressible noise; i.e. descriptions and non-random searches are both pointless

83. #83 snaxalotl
May 16, 2009

“…successful searches must in some sense encode information about the landscapes that they search”

remember that in evolution the landscape is part of the search mechanism

84. #84 Erik 12345
May 17, 2009

@79: How do you prove it using the NFL theorem?

85. #85 isotelesis
May 19, 2009

The multiverse is basically another version of Darwinism, soon they’ll call it the omniverse, perhaps even a Zionverse.

IS THEORETICAL PHYSICS BECOMING THE NEXT BATTLEGROUND IN THE CULTURE WARS? NOT ACCORDING TO SOME THEOLOGIANS AND SCIENTISTS.

http://seedmagazine.com/content/article/the_multiverse_problem/

86. #86 Marcel Kincaid
May 20, 2009

All real numbers can be represented as limits of rational numbers

Ignorant meaningless gibberish.

(or as infinite decimal expansions, if you wish)

Go ahead and give us an infinite decimal expansion of a random transcendental.

That not all real numbers can be represented using, say, finite decimal expansions has more to do with “size” (cardinality) of the set of real numbers than the NFL theorem.

As Stephen Wells points out in effect, this is a fallacy of affirmation of the consequent. It’s also a much weaker claim than Mark’s: “The numbers that we can represent in any notation are a miniscule subset of the set of all real numbers.”

How do you prove it using the NFL theorem?

A nice bit of goalpost moving and burden shifting. You wrote “the connection claimed here between (finite?) representations of real numbers and the NFL theorem is surely not correct”. Aside from the mischaracterization of what Mark wrote, go ahead and prove that. After you fail miserably to do so, be prepared to be embarrassed when Mark gives you the proof he claimed exists and that you have no reason to doubt. (Hint: think of a mapping between searches of the real number space and notations.)

87. #87 Erik 12345
May 21, 2009

Marcel Kincaid, real numbers can indeed be represented as limits of a sequences of rational numbers. An infinite decimal expansion is one such limit.

If you have a proof that makes use of the NFL theorem, please give it. Such a proof does not appear to exists, since the NFL theorem deals only with performance averaged over all objective/fitness/cost functions.

Your hint of thinking of a mapping between “searches” of the real numbers and notations is not very helpful. It is unclear what mapping you have in mind. The paper on the NFL theorem made use of a few concepts like: search space, set of possible cost values, cost function (or “fitness function”), search algorithm. Are you suggesting a mapping between notations and search algorithms? Where do cost functions and averages over all possible cost functions come in?

88. #88 Jonathan Vos Post
May 21, 2009

Sir Francis Crick, after sharing the Nobel Prize for the the Double Helix research, spent 20 years studying consciousness. He an various coauthors, particularly Cristof Koch, wrote about NCC: the Neural Correlates of Consciouness. The considered whether or not there was a minimal level of complexity necessary for a platform to support consciousness. Granted, consciousness is not intelligence. One can have intelligence without consciousness. One distinguishes states of consciousness from content of consciousness. The Intelligent Design liars have many hidden assumptions about their putative Intelligent Designer. He he/she/it conscious? Beats me.

This comment abstracts a 47-page draft paper based on two talks given at Caltech last weekend, by Cristof Koch and Barbara Wold. I think it connects with the Dembski Deconstruction, via attempting to define “intelligent” and “Complexity.”

Your gut contains roughly a billion neurons, about 1% as many as in your brain and spinal cord. Hence your Enteric Nervous System is sometimes called your “second brain.”

My dog’s brain contains about as many neurons as my “second brain.” I strongly believe that my dog is conscious. Other intelligent people deny consciousness to dogs, cat5s, even chimps and dolphins. So why is there no such debate as to whether or not your gut is conscious?

The human immune system learns. It is as complex as some neural networks. Your genome is a complex network, whose circuitry is being increasingly understood. Is the complexity of the human immune system or of the human gene network above the threshold of complexity prerequisite for consciousness?

Koch and Wold agreed that neither question is frivolous. They gave interesting partial answers, which the margin of this comment is too small to contain.

89. #89 DarkKnight
May 22, 2009

Macro Evolution is not logically possible, take the the
analogy of Pirated Software, imagine Pirated software is copied and passed on only if its a working crack, occasionally you get corrupted copies, which do not get passed around.

Macro evolution suggests that if the corruption causes a ‘good’ change then it becomes more popular, more people have copies of it and the older versions die out. Now iterate this process,this is macro evolution in a nutshell and we are told that a notepad application can somehow transform into a something like MATLAB..

90. #90 Skemono
May 24, 2009

Instead of coming up with crappy analogies that “disprove” evolution, how about you try dealing with the actual evidence out there? The fossil record, DNA evidence, etc.

91. #91 Ray S.
June 7, 2009

DarkKnight @ 86

Your analogy of pirated cracked software doesn’t seem to reflect the genetic activity of living generations, nor does it take into account the time scales involved. Some of us see no reason a properly constructed GA might not develop MATLAB from notepad. Your claim thus boils down to an argument from personal incredulity.

92. #92 Howard C. Shaw III
August 21, 2009

DarkKnight @ 86, Ray S. @ 90

Or some people might just point to Emacs and say, ‘it already has’. http://members3.jcom.home.ne.jp/imaxima/Site/Welcome.html

Heck, not only did a notepad-alike evolve into a MATLAB-alike while being passed around, it also became an integrated development environment, and for some people, an operating system. And for others, *snicker*, a religion.
http://www.dina.dk/~abraham/religion/

Though, granted, it was not pirated.

Howard C. Shaw III