Those pesky ‘pathetic’ details…

| 44 Comments

Dembski commented on a discussion on the BBC between himself and Ken Miller and attempts to address two issues raised by Ken Miller. As I shall show, in both cases Dembski fails.

Dembski Wrote:

(1) The main weakness of evolution is that it is science (yes, Miller actually did say this and went on so long about it that the BBC host could not give me my closing comment as he had intended to) and (2) ID’s main fault is that it proceeds by negative argumentation.

This is going to be interesting. Dembski, who is a philosopher, mathematician and theologian may not be too familiar with the scientific evidence for evolution. As far as how he is going to argue against (2) is beyond me. It’s self evident that ID’s approach is by negative argumentation. In fact when Dembski was asked for details for Intelligent Design he responded

Dembski Wrote:

“You’re asking me to play a game: ‘Provide as much detail in terms of possible causal mechanisms for your ID position as I do for my Darwinian position.’ ID is not a mechanistic theory, and it’s not ID’s task to match your pathetic level of detail in telling mechanistic stories.”

So what is Dembski’s argument about evolution and science?

Dembski Wrote:

Point (1): The problem is not that evolution is science and therefore rife with open problems. The problem is that it never solved the problem which it set itself, namely the increasing complexification of life over the course of natural history. To be sure, evolutionists claim to have solved this problem by uncovering natural selection and other material mechanisms. But evolutionists have never been able to take these mechanisms and with them provide detailed, testable accounts of how the origination of complex biological systems required for macroevolution could have occurred. Plenty of handwaving, yes; details that would convince and evolution skeptic, no.

However, the bar set by Dembski is not only unreasonably high but while evolution has at least proposed mechanisms and pathways which can be tested, ID basically has nothing more than ‘poof’. In fact, science has proposed many pathways and mechanisms to explain complexity in life. The problem is that Dembski’s definition of complexity is basically the log of the probability, or in other words, something is complex when it cannot be explained by chance or regularity. The moment one explains ‘complexity’ it is not complex anymore. Complexity, as used in science typically looks at other aspects and Dembski conflates the various meanings leading to confusion amongst his faithful as well as amongst unaware citizens and even scientists.

Judge Jones was very critical of this

Judge Jones Wrote:

It is appropriate at this juncture to address ID’s claims against evolution. ID proponents support their assertion that evolutionary theory cannot account for life’s complexity by pointing to real gaps in scientific knowledge, which indisputably exist in all scientific theories, but also by misrepresenting wellestablished scientific propositions. (1:112, 1:122, 1:136-37 (Miller); 16:74-79, 17:45-46 (Padian)).

And is Dembski right, has science failed to explain the complexification of life? Has science failed to explain how information can increase in the genome?

On the contrary, people like Tom Schneider or Chris Adami et al have shown how the simple processes of mutation and selection can in fact increase the information in the genome.

T. D. Schneider, Evolution of Biological Information, Nucleic Acids Res, 28, 14, 2794-2799, 2000

Adami, Ofria, Collier Evolution of biological complexity Proc. Nat. Acad. Sci (USA) 97 (2000) 4463

And let’s not forget the seminal work by Lenski et al on the evolution of complex features.

Lenski RE, Ofria C, Pennock RT, and Adami C The Evolutionary Origin of Complex Features Nature, 423:139-144 (2003).

And the work on self organization, self organized criticality, gene networks, and evolvability show countless papers and examples of how science is dealing with these issues. What has ID contributed to this discussion? Poof…

Or lets consider the combination of evolution and development which has led to many more insights into mechanisms and pathways of evolution. Intelligent Design’s Icon, the Cambrian Explosion was once argued to be an example of independent origins. Now that science has progressed and more data becomes available, it is clear that the Cambrian explosion is well explainable in Darwinian terms. James Valentine, often quoted by ID proponents, and a foremost expert on the Cambrian explosion has reached this conclusion.

So let’s visit Dembski’s second ‘argument’ and see if he does any better here.

Dembski Wrote:

Point (2): Negative argumentation for one of two mutually exclusive and exhaustive positions is always positive argumentation for the other (the two positions here are intelligent design and unintelligent evolution, i.e., evolution that proceeds without intelligent input). Yes, much of ID argumentation is showing the limits to evolvability of various biological systems given certain material mechanisms. But the charge of negative argumentation applies equally to evolutionary theory: evolution argues negatively against ID. Just as ID hasn’t ruled out all conceivable material mechanisms, evolution has not ruled out all conceivable actions by intelligent agents in forming biological complexity. ID has this advantage, however. We do know that intelligent agents can bring about the types of functional systems we see in biology; we have no evidence that unintelligent evolution can do the same.

This points to a common flaw in Dembski’s argument namely that it is an either or proposition. But as Dembski has argued himself even if evolution happened fully by natural means, it could still have involved (but not necessarily required) an intelligent input. Secondly, Dembski argues that even if Design happened, it may very well be that it can also be explained by natural processes. In fact, unless there is some positive evidence, the approach chosen by Dembski cannot even compete with ‘we don’t know’. That’s because Dembski’s ‘Design Inference’ is basically nothing more than the null hypothesis. There are so many problems with Dembski’s claim here that it is clear that Dembski is approaching this issue from a purely mathematical or philosophical perspective but when it comes to dealing with it from a scientific perspective his arguments quickly are shown to be irrelevant or plain wrong. Let’s for instance look at Dembski’s claim that the Design Inference is free from false positives. Now in theory, this is correct, if and only if all known and unknown hypotheses for chance and regularity have been rejected can one reliably infer not(chance or regularity). But this requires a level of knowledge typically not available to human intelligence. In most cases, especially in the sciences, the situation is not that black or white. Incomplete knowledge, incomplete understanding of the probabilities involved all will lead to inevitable false positives (instances in which design was incorrectly inferred). History has shown us many instances where our ignorance led us to infer ‘Intelligent Design’, only to later be forced to accept a far less reaching conclusion.

So while Dembski originally argued that false positives would render his approach useless, he admitted over time, that false positives were indeed a real possibility. In other words, Dembski himself has come to the inevitable conclusion that his Design Inference is ‘useless’.

See for instance my more in-depth postings on this topic. Icons of ID: Reliability revisited and Icons of ID: Explanatory filter and false positives.

Science however does not rely on disproving all known possible actions by intelligent design agents, and in fact accepts that the intelligent designer proposed by ID, namely God,can never be ruled out. While Dembski argues that evidence in favor of evolution argues against intelligent design, his arguments are illogical and even contradictory to his own arguments. In fact, Judge Jones addressed this confusion when pointing out that evolution and “intelligent design” can co-exist quite peacefully. Exactly because they are NOT mutually exclusive.

Judge Jones Wrote:

Both Defendants and many of the leading proponents of ID make a bedrock assumption which is utterly false. Their presupposition is that evolutionary theory is antithetical to a belief in the existence of a supreme being and to religion in general. Repeatedly in this trial, Plaintiffs’ scientific experts testified that the theory of evolution represents good science, is overwhelmingly accepted by the scientific community, and that it in no way conflicts with, nor does it deny, the existence of a divine creator.

So let’s attempt a little thought experiment that may clarify my position here:

Ask this question of yourself: Would you reject your belief in God just because science has shown that God is not involved in the continuous manipulation of the bodies rotating our sun? Is there really ANY scientific evidence which could lead you to reject your faith in God?

Dembski himself has argued that natural law and chance may very well fully explain complexity in our universe, leaving the origin of such information outside the scope of science.

So how does science deal with this? And why does it deal with it much better than Intelligent Design? Again the answer is trivially simpl: science proposes positive, and testable hypotheses which are consequently subjected to attempts to disprove. The more such hypotheses resist disproof and the more evidence supporting these hypotheses are found, the better these hypotheses will do. Eventually, the evidence may be such that these hypotheses become theories. For instance, in case of evolution, the evidence for Common Descent has been so succesful that by all measures of science, it is a scientific fact. However, how one explains Common Descent, remains an issue of theory and hypotheses.

Note that Intelligent Design does not propose ANY scientifically relevant hypotheses nor makes ANY claim about how to explain a particular feature while science proposes plausible pathways and mechanisms which may thus be subjected to disproof.

Which is why many scientists have come to the conclusion that Intelligent Design is scientifically vacuous. But this conclusion is not limited to scientists, also many religious people have come to a similar conclusion.

Now, of course when one is convinced about the existence of the Intelligent Designer, one may argue that His interactions with His Creation must be detectable. And yet, even Dembski has come to realize that the Designer may have been fully ‘hands-off’ letting chance and regularity processes reign. Such a position would clearly reconcile faith (Intelligent Design) and science as it would place the original action of the Intelligent Designer outside the realm of scientific inquiry. For example, by displacing the origin of complex specified information to a moment before the Planck Time, one can reconcile the scientific facts with the possibility of Design. Or one could argue that a Deity interacts with His Creation via purely natural processes, making such interactions indistinguishable from natural processes.

Dembski Wrote:

There is no logical contradiction here. Nor is there necessarily a god-of-the-gaps problem here. It’s certainly conceivable that a supernatural agent could act in the world by moving particles so that the resulting discontinuity in the chain of physical causality could never be removed by appealing to purely physical forces. The “gaps” in the god-of-the-gaps objection are meant to denote gaps of ignorance about underlying physical mechanisms. But there’s no reason to think that all gaps must give way to ordinary physical explanations once we know enough about the underlying physical mechanisms. The mechanisms may simply not exist. Some gaps might constitute ontic discontinuities in the chain of physical causes and thus remain forever beyond the capacity of physical mechanisms.

Link

Dembski does not like this ‘solution’ and thus looks for a different way for the unembodied designer to interact with His Creation.

Now Dembski has tried to resolve this major problem by proposing that such a unembodied designer may transmit information to His Creation via an infinite wavelength.

Dembski Wrote:

How much energy is required to impart information? We have sensors that can detect quantum events and amplify them to the macroscopic level. What’s more, the energy in quantum events is proportional to frequency or inversely proportional to wavelength. And since there is no upper limit to the wavelength of, for instance, electromagnetic radiation, there is no lower limit to the energy required to impart information. In the limit, a designer could therefore impart information into the universe without inputting any energy at all.

While mathematically and philosophically interesting since this would make the energy of the signal zero, a scientist would quickly realize the follies of such a proposal. While ID has some interesting philosophical implications, these examples also show how quickly ID becomes scientifically vacuous.

Now, Dembski has argued in the past that complex specified information is a clear indicator of Intelligent Design but as Wesley Elsberry so cleverly showed, Dembski had to admit that algorithms can in fact create complex specified information and concluded that there is such a thing as apparant versus actual complex specified information. Dembski has yet to address this major problem in his logic.

Wesley Elsberry Wrote:

If Dembski’s analytical techniques cannot resolve the issue of possible cheating in the “Algorithm Room”, how does he hope to resolve the issue of whether certain features of biology are necessarily the work of an intelligent agent or agents? If Dembski has no solution to this dilemma, the Design Inference is dead.

So in other words: The Design Inference is by virtue of being the set theoretic complement of chance and regularity (also known as appeal from ignorance) inevitably subject to false positives. Complex Specified Information, once touted as a reliable indicator of Intelligent Design, now has two forms “apparant” and “actual”, where the determination of “apparant” is made when it involves natural processes. Of course, how one is to resolve “apparant” versus “actual” is yet another unexplained issue.

There is far more wrong with Dembski’s arguments and I could spend hours discussing the work by Mark Perakh, Wesley Elsberry, Jeffrey Shallit, John Wilkins, Elliot Sober and many others who have shown from various perspectives why Intelligent Design Fails but this seems sufficient to lay to rest Dembski’s confused claims about science and intelligent design.

On UncommonDescent Pmob1 obseves that (since Dembski tends to remove comments which are critical of his position, I have copied the comment below)

D writes: “Negative argumentation for one of two mutually exclusive and exhaustive positions is always positive argumentation for the other (the two positions here are intelligent design and unintelligent evolution…”

I think D assumes far too much in this proposition. I can imagine designed body forms that subsequently develop haphazardly in nature. I can imagine heritable alterations of DNA caused by cellular, extra-cellular or environmental events. I can entertain the idea of speciation of designed forms.

I don’t claim to have empirical evidence of the above. Situation: contingent. State of knowledge: fluid. But please note that D’s proposition implies that he DOES claim empirical evidence, i.e., in his use of the term “exhaustive.” Clearly D has no such exhaustive evidence. Therefore his claim of mutual exclusivity is false. D describes a rhetorical, not a logical, battle when referencing negative argumentations in this instance.

44 Comments

Actually I’d disagree that ID is based on negative argumentation. As far as I can tell it seems to be based on positive maths-based argumentation. Sadly the maths in question is dodgy, so it doesn’t work, but still.

Actually I’d disagree that ID is based on negative argumentation

as John Stuart would say;

“Whaaaa?”

Actually I’d disagree that ID is based on negative argumentation. As far as I can tell it seems to be based on positive maths-based argumentation. Sadly the maths in question is dodgy, so it doesn’t work, but still.

Which feeble IDC argument would you be referring to? Behe’s ‘irreducible complexity’ is undoubtedly a negative argument: “insufficient evidence for evolution through natural means, therefore it must be ID”. Fancifying the rhetoric to make it sound like a positive argument does not change the underlying logic. No one has produced evidence of any ‘design event’. No one has produced a potential mechanism for any ‘design event’ beyond Behe’s “poof and a cloud of smoke”.

Corkscrew Wrote:

Actually I’d disagree that ID is based on negative argumentation. As far as I can tell it seems to be based on positive maths-based argumentation. Sadly the maths in question is dodgy, so it doesn’t work, but still.

ID is based on the set theoretic complement of chance and regularity or in other words NOT(Chance and/or regularity) THUS design.

I’d like to hear why you believe that it is not based on negative argumentation

Corkscrew, the problem is that even the faulty math-based argumentation of Dembski is at heart negative, not positive. The argument is always “evolution could not have done this because” or “a natural process could not have done this because”. The argument is never “we know the intelligent designer did this because”.

Dembski is quite skilled at taking a simple idea and using so many words to describe it that by the end it seems, at a glance, to be something quite other than what it is; but if you will go back and look at that dodgy math again, and keep a careful eye on the ideas at the core of all the philosophizing this time, you will find that even in a mathematical sense it is always negative, never positive. The fallacy of false choice is as deeply woven into the design inference as it could possibly be.

I agree. The argument is a negative argument, but not a factually correct negative argument. I think Miller gives them too much credit by stating that they have a negative argument at all. Is there actual data that demonstrates that RM + NS can’t create diversity ? No. They ASSERT that RM + NS can’t (a negative argument in principle), but that doesn’t make it factually true.

While reading this post and the comments, I was thinking about the appeal of binary thinking (i.e., there are only two sides to an argument; something is either on or it’s off). I’ve read about studies that demonstrate how high different species can count, and I’ve long heard the conventional wisdom that we have a 10-base number system because we have 10 fingers. But no matter what the human brain is capable of in the way of counting or how many fingers we have, there seems to be an attraction to the “on the one hand, on the other hand” kind of argument. Maybe it’s because we have two hands!?! Binary sorting is a very practical tool. I use whenever I have to alphabetize a stack of cards or papers on my lap. But it is a multi-step process. You can’t just stack the cards into two piles and be done with it; you need to do progressive sorts of each pile until everything is in the proper order. The simplicity of the either-or proposition is definitely appealling to people. The right wing knows this and attacks anything that does not adhere to this way of thinking as “cultural relativism.” The only question I have is: do Dembski and his fellow IDers actually lack the capacity to move to the next step in the binary thought process, or are they cynically exploiting the appeal of a simple either/or argument?

PvM asked:

I’d like to hear why you believe that it is not based on negative argumentation

Doesn’t Dembski think he has a positive case for detecting design and therefore intelligence?

There are two ways to deal with that claim: 1) Claim Dembski hasn’t detected intelligence or 2) Say he has defined intelligence of a sort with his scheme, but that evolution is intelligent, in that way.

It all hinges on defining intelligence.

Why not say that evolution has the intelligence required – it records the imperfect memory of a billion plus years of life experience and learns something from that experience. Thus evolutionary algorithms are a mechanism of intelligence and in fact genetic and evolutionary algorithms are used in artificial intelligence research.

Doesn’t Dembski think he has a positive case for detecting design and therefore intelligence?

Dembski thinks a number of strange things. I’m not sure he’s demonstrated himself capable to objectively evaluate the nature of his own arguments.

and I’ve long heard the conventional wisdom that we have a 10-base number system because we have 10 fingers.

Except for those cultures which have a base 20 numbering system because they have fingers and toes.

and I’ve long heard the conventional wisdom that we have a 10-base number system because we have 10 fingers.

Except for those cultures which have a base 20 numbering system because they have fingers and toes.

What does this say about the cultures with a base 60 numbering system?

:O

Actually I’d disagree that ID is based on negative argumentation. As far as I can tell it seems to be based on positive maths-based argumentation. Sadly the maths in question is dodgy, so it doesn’t work, but still.

Alas, Dembski’s math is all just a fancy argument for “god of the gaps”. Boiled down to its essentials, Dembski’s argument is “if not chance, if not law, then design”. I.e., “if we can’t explain it, it must be designed”. I.e., god of the gaps.

I have often asked Dembski apologists why the, uh, Isaac Newton of Information Theory doesn’t switch his formula round to say, “If not design, if not law, then chance” or “if not chance, if not design, then law”. Answer? Because the forumula he puts forth is the only one that allows Dembski to avoid making ANY prediction, test or evaluation for “design”.

I suspect that is not a coincidence.

It seems possible that Judge Jones has put his finger on what makes Dembski uncomfortable. The theistic evolutionists are essentially saying “According to our best understanding of the evidence currently available, evolution works by the mechanisms specified, AND one or more gods exist.” Jones points out that this is perfectly consistent. However, that last part should be barred from the classroom because it promotes religion and serves no secular purpose - i.e. it adds nothing explanatory whatsoever. And Dembski seems to notice that if his particular god is omitted and nothing is lost, then his god isn’t getting the credit Dembski would prefer. Dembski would prefer that the FIRST part of the statement be omitted, on the grounds that if it is NOT omitted, then people are more likely to notice that it’s sufficient.

and I’ve long heard the conventional wisdom that we have a 10-base number system because we have 10 fingers.

Except for those cultures which have a base 20 numbering system because they have fingers and toes.

I wished they’d have skipped the thumbs and big toes. Then we’d have a hexadecimal system - much more convenient.

PvM wrote: “Now, Dembski has argued in the past that complex specified information is a clear indicator of Intelligent Design but as Wesley Elsberry so cleverly showed, Dembski had to admit that algorithms can in fact create complex specified information and concluded that there is such a thing as apparant versus actual complex specified information. Dembski has yet to address this major problem in his logic.”

Doesn’t Dembski simply argue that the algorithm was itself designed? That the CSI is slipped, under the table, into the algorithm by an intelligent agent. I know this is false, but I have yet to find a convincing way to demonstrate that natural selection is a perfectly legitimate way to filter life to the people I talk to. I show them a genetic algorithm and they come back with Demsbski’s “you slipped the CSI into the algorithm”. Since I am trying to compare the genetic algorithm with natural selection they say that natural selection is therefore artificially (i.e. intelligently) constrained. AAARRRRGGGGGHHHHHH!!!!!!! It’s like arguing with a greased pig.

Sincerely, Paul

P.S. DOUBLE ARRGGGHHHH!!!!! I had a guy come into my bookstore Friday night specifically looking for a book by Dembski. I wound up selling him the copy of NFL I ordered some months back and hadn’t yet done more than skim because of the holiday season. Not that I mind getting rid of the book, but why do I have to participate in enriching this snake-oil quack?

I wished they’d have skipped the thumbs and big toes. Then we’d have a hexadecimal system - much more convenient.

Not so great for representing fractions, though.

I’m a big fan of base 12. It’s a good compromise.

“Actually I’d disagree that ID is based on negative argumentation.”

as John Stuart would say;

“Whaaaa?”

The version that I heard stated (roughly speaking) that there was some kind of conserved quantity (mis)labelled “information”* that was conserved with respect to natural processes but could be generated by intelligence (I’m paraphrasing here, I hope I’ve got the gist of it). I’d say that the detection of such “information” could then be considered a positive argument in favour of the conjecture that an intelligent entity diddit, in the same way that a rise in temperature of a closed system would comprise positive evidence of a heater of some sort**.

Shame it all falls to pieces when you note that neither their variant, nor any other variant (that I’m aware of), of information is actually conserved.

*I will continue to put quotes round this word ad nauseam because my exams this year include problems related to Shannon information and I don’t want to get confused.

** Yes, I’m aware that, for example, water going over a waterfall warms up slightly. No analogy is perfect.

Paul Flocken wrote

. Since I am trying to compare the genetic algorithm with natural selection they say that natural selection is therefore artificially (i.e. intelligently) constrained.

This can be addressed by bringing up other examples where algorithms are used to model a natural system that IDers don’t have any particular problem with - orbital mechanics, say. Ask how a genetic algorithm has their magic intelligence in it while Kepler doesn’t, when they’re both written by humans applying the same techniques and with the same skill set. It quickly boils down to ID = Heap Big Magic You Can’t See, or an exercise in wilful misunderstanding.

R

Corkscrew, do you know of any definition of information that the proponents of that argument have? Because I cannot think of a good definition that fits the conservation requirement they’ve presented you.

Let’s define information as the lack of patterns. That is, the more regular a system is, the easier it is to describe in terms of simple formula, and you can make more generalizations about the system without losing the ability to describe it exactly. A low-information system would be a cubical grid of identical particles who are all traveling at the same velocity. If you know the information of one particle (how big it is, how fast it’s going, where it’s going, where it is in the grid), you can describe all of the other particles from that basis, and you know everything about the particles. I’d call that a low-information system. Just for kicks, let’s make it a closed system, with impenetrable walls, so no information can get in. Can we agree that this is low-information? Assuming we can, I will continue.

Our particles will not stay predictable for very long. Inevitably, they will attract or repel each other, or bounce off the walls, and things will go downhill. Even with the simplest particles, the interactions are nondeterministic, and random changes soon fill our system. Soon, our particles have all departed from their original course, and we have to define a unique position and velocity for each of them. Furthermore, even if we only have four particles, any interactive forces between them will make the system atrociously difficult to model, and it gets worse for every particle you add after that. And if the particles are quantum particles, you will soon be up past your neck information. Our ‘low-information’ system isn’t anymore.

A small system of a hundred particles can never be modeled exactly, because there’s too much information. A system like a cell, which contains trillions of such particles, is so full of information that we can never even hope to be able to describe every last detail. A system like ours, i.e. a planet covered with biological systems and heated by a star, is millions of orders of magnitude worse. It’s so complex that it’s utterly unpredictable. The ID advocates had better have a better definition of ‘information’, because by the previous definition (which I see as a good one), the information present in life is too ridiculously huge to be measured, and any increase in it could just be due to random interactions.

Incidentally, I just argued that random interactions not only can produce more information in a closed system, but that they do all the time and that they produce quite a lot. Isn’t that what various pro-evolution scientists have been trying to argue - that the sum of random changes can easily produce more complexity in a system?

Tice with a J wrote:

Let’s define information as the lack of patterns.

No, let’s not. That would be distorting and dishonest.

In some ways that’s the opposite of what Claude Shannon and Kolmogorov had in mind – the less pattern the more information in it. The more complexity, the more information. And sometimes we don’t see the patterns that are there.

Information is a term with many meanings depending on context, but is as a rule closely related to such concepts as meaning, knowledge, instruction, communication, representation, and mental stimulus.

Information is what human beings try to do with data.

Information might be defined as a message received and understood that reduces the recipient’s uncertainty. Or it might be defined as a collection of facts from which conclusions may be drawn. In communication theory it’s a numerical measure of the uncertainty of an outcome.

There is only one context in which “information” can be associated with “pattern” and that’s in neural net theory. Now you put it in that context and you will completely undermine ID because as you note:

…random interactions not only can produce more information in a closed system, but that they do all the time and that they produce quite a lot. Isn’t that what various pro-evolution scientists have been trying to argue - that the sum of random changes can easily produce more complexity in a system?

Howeverm to us human beings, in general, the more raw complexity and data needed, the less useful the information seems to be.

Lets just say all our ideas about information (as opposed to data) are anthropocentric and cannot be generalized without confusing people.

Tice with a J Wrote:

Corkscrew, do you know of any definition of information that the proponents of that argument have? Because I cannot think of a good definition that fits the conservation requirement they’ve presented you.

See here for a brief summary. The site owner picks up very effectively on what I consider to be the elephant in the CSI room.

Incidentally, I strongly recommend sticking to the Shannon definition of information if you feel the need to define it. Firstly because it prevents confusion among those with a clue, and secondly because it raises the bar beyond the reach of those who are only pretending to have a clue, which is an effective way of driving them blinking into the daylight.

Excellent points, Norman and Corkscrew. I will take your advice. In my original post, I was partly working from the definition of physical information, but mostly from off the top of my head. However, in trying to come up with a definition of information that isn’t anthropocentric, I think I could have done worse. Of course, all that this talk of ‘information’ really does is show how silly it is trying to formulate an origins theory around information, as Dembski tried to do.

By the way, I have to say that when I see CSI, I think Crime Scene Investigator. Seeing it stated here so frequently is making me giggle.

Paul Flocken Wrote:

Since I am trying to compare the genetic algorithm with natural selection they say that natural selection is therefore artificially (i.e. intelligently) constrained.

The only reason genetic algorithms have any ‘specified complexity’ to them is that they have no natural fitness function. You have to introduce an artificial means to say what’s better and by how much.

The whole reason natural selection doesn’t require such formulae is because nature has a natural mix based on energy, resources, predation, etc. Nature doesn’t care whether 15-body gravity equations are “in the realm of impossibility to solve”. Growing a large body requires a lot of food, and the capability to digest it. You don’t see a lot of deep ocean monkeys scooping plankton, or spinach-eaters suffering from pellagra. There are peaks and valleys in this “massively differential equation” that shift over time. You can simulate it. It might seem like a lot of “information” in that simulation, but the conditions can be shifted and somewhat randomized without much effect (randomizing a lot gives you a disaster scenario, and those happen, too)

I wished they’d have skipped the thumbs and big toes. Then we’d have a hexadecimal system - much more convenient.

Hey! I can see what you’re trying to do! That’s just binary in a cheap tuxedo!

Or should that be 4 cheap tuxedos?

Bob

In related news dembski gives up…

Mothballing Uncommon Descent

I’ve decided to put Uncommon Descent into mothballs indefinitely. Although I’ve enjoyed blogging, I find it distracts from more pressing work that I need to get done.

Tice with a J Wrote:

However, in trying to come up with a definition of information that isn’t anthropocentric, I think I could have done worse.

Yeah, your definition is Not Bad and is in fact pretty much the definition of Kolmogorov complexity (another oft-discussed topic in ID circles). The problem with this form of information-like quantity is that it is formally impossible to produce a function that will compute it. This is because of nasty little stings in the K-complex tail like the Berry paradox, which IIRC goes something like: “Consider the shortest K-complex string that can’t be defined in twenty words or less. Well, I’ve just defined it in twenty words or less.”

Have you come across the Shannon definition of information before? Would it be useful to you to have an intuitive definition?

Flint Wrote:

The theistic evolutionists are essentially saying “According to our best understanding of the evidence currently available, evolution works by the mechanisms specified, AND one or more gods exist.” Jones points out that this is perfectly consistent. However, that last part should be barred from the classroom because it promotes religion and serves no secular purpose - i.e. it adds nothing explanatory whatsoever.…

Since, as you notice, the “god(s) exist” part adds nothing to the explanation, it is incorrect to say that belief in those god(s) is based on evidence at all, so any theistic evolutionist making that claim would be mistaken or dishonest. It would be more honest to admit the “belief in god(s)” part is based on faith, not evidence.

If you or anyone could produce reliable scientific evidence that god(s) exist, I would have no objection whatsover to it being taught in a public school science classroom.

When reading the dialog with Dembski, I am reminded of a joke that was included in Isaac Asimov’s Treasury of Humor. In it, a college president is complaining to the head of the physics department about the expensive hardware his scientists use. He then says: “Look at the math department! All they need is chalk and erasers! And the philosophy department only needs chalk!”

People like Dembski will *never* admit that they are wrong. They seem to have no concept of shame or objective integrity.

The philosopher Mario Bunge somewhere (I think in his Philosophy of Psychology) lists a dozen or something different uses of “information” and suggests a lot of problems can be solved by recognizing that they aren’t at all equivalent understanding.

If Dembski had a valid negative argument, that would be remarkable enough, though it still wouldn’t prove the existence of an intelligent designer. By a negative argument, I mean (since he’s allegedly a mathematician) a mathematical proof that something is not possible–namely, the generation of a functioning system of a certain “complexity” (in some rigorous sense to be determined) with reasonable probability as the result of an evolutionary process. It’s conceivable that such things can be proved in a vaguely analogous sense to how it can be proved that an angle cannot be trisected with compass and straightedge or that a computer program cannot solve the halting problem.

On the other hand, to see how difficult it is, note that a lot of problems that look like finding a “design” to fulfill a task can be reduced to problems in NP, co-NP, and a few levels higher up in the polynomial time hierarchy (e.g. find a minimal chip layout of a fast digital logic circuit that calculates a truth table specified by some particular combinational logic expression–not in NP, but I think in the union of NP and co-NP). There is currently not even a mathematical proof that these “intractable” optimization problems are not in P (e.g. solvable using a deterministic polynomial time algorithm). It’s just hard for me to imagine how, if you cannot even disprove the existence of a polynomial time digital computer program for generating “design” that you can be expected to prove that a complex system running over the universe as a whole (including effects not obviously reducible to digital logic by the way) cannot produce “design.”

The other thing about Dembski is he must know that he does not have a single valid argument, because he periodically changes his arguments without ever quite finishing them. If he picked one valid negative attack on evolution and carried it to completion, that would be astonishing (particularly, IMHO, that there isn’t likely to be one). Removed from the context of biology, it would still be a result of interest to mathematicians and would win him awards rather than his current pariah status.

This whole argument about information and the impossibility of complexity arising from a less complex state reminds me a lot of the old creationist tactic of saying that the evolution of life on the planet violates the laws of thermodynamics, in that the entropy of the planet is (at least initially) decreasing with time in an evolutionary model.

I also remember some time ago reading a paper that related the degree of entropy of a system to the total amount of information it can encode/complexity it has. Does this make sense? I’m not an information theorist, just a lowly chemist, so I don’t know if this argument is sound. But if it is, it would make for a very elegant refutation of Dembski’s thesis, since the framework for refuting the entropy argument already exists and is easily provable.

uberhobo,

Unfortunately, although I know about Thermodynamics I know nothing about information theory.

My *guess* is that higher entropy contains more information. Lower entropy contains less information.

I’d just like to point out to Lenny Flank that Dembski was only called the “Isaac Newton of Information Theory” because Newton was involved in pointless attempts at Alchemy.

Syntax Error: not well-formed (invalid token) at line 7, column 7, byte 248 at /usr/local/lib/perl5/site_perl/5.12.3/mach/XML/Parser.pm line 187

He’s making a killing off of it while laughing at both his detractors and his supporters.

He’s entirely welcome to laugh all the way home from those science classrooms that he’s not getting into. (shrug)

Comment #65147

Posted by jim on December 27, 2005 02:35 PM (e) (s)

uberhobo,

Unfortunately, although I know about Thermodynamics I know nothing about information theory.

My *guess* is that higher entropy contains more information. Lower entropy contains less information.

I think Shannon-Weaver theory says maximal noise is maximal information, but that’s a vague memory.

In any case, information theory is worthless to the ID Creationists because they never managed to find a kind of ‘information’ which can only be produced by a mind. The whole point of the “specified” in complex specified information was to try to do this, but they could never make it work.

* And finally, I would cash in on the celebrity associated with bringing down Darwin.

On this last point, let me just say that intelligent design has been very, very good to me.

The irony is that the poor, deluded ID fans couldn’t believe something like this, where I recognize that most of the people like Demski, Falwell, Robertson may have started out honestly, but in the end Mammon gets them everytime. I guess that’s from having too many clients in the Christian music business and watching how it works as a confidant.

Before you take that “Why I want to bring down evolution” confession as some kind of evidence on dembski, be aware he posted it on “April the 1st”. The sad thing is most of it is still probably true.

Syntax Error: mismatched tag at line 20, column 2, byte 1338 at /usr/local/lib/perl5/site_perl/5.12.3/mach/XML/Parser.pm line 187

not for nothing, but it reminds me of his “Street Theatre” performance after he was caught in several lies.

using it on april 1st just gives him yet another escape route.

I’t more likely he made that up as a cover story for when things aren’t going so well for him… like now.

what’s your support for that assumption, you ask? take a look at the wording in his own statement:

Many critics of intelligent design think I got into ID because I’m a right wing fundamentalist who takes Genesis literally and wants to see its 6-day creation as mandatory teaching in the public school science curriculum. WRONG! Here’s the real story.

Back in the mid-80s, Sahotra Sarkar and I were grad students at the University of Chicago, he in philosophy, I in mathematics. One day, at the International House (a 500-room dorm for grad students on the UofC campus), we were discussing what it would take to bring down Darwinism.

one, anybody who capitalizes the word “wrong” is overcompensating.

two, when someone refers to evolutionary theory as “Darwinism”, I don’t believe for a second they have a firm grasp of the theory to begin with.

I’d be willing to be that the thing he is “having us on” about is that his father teaches evolutionary biology.

sounds like he threw that in there as a lame attempt at an argument from authority, to me.

Posted by sir_toejam on December 29, 2005 01:20 AM (e) (s)

two, when someone refers to evolutionary theory as “Darwinism”, I don’t believe for a second they have a firm grasp of the theory to begin with…

Possibly. Alternatively he could have a firm grasp and using the term “Darwinism” might lead to other implications. Such as political and or religious motives.

Then again, it could be all of the above.

Many critics of intelligent design think I got into ID because I’m a right wing fundamentalist who takes Genesis literally and wants to see its 6-day creation as mandatory teaching in the public school science curriculum. WRONG! Here’s the real story.

Back in the mid-80s, Sahotra Sarkar and I were grad students at the University of Chicago, he in philosophy, I in mathematics. One day, at the International House (a 500-room dorm for grad students on the UofC campus), we were discussing what it would take to bring down Darwinism.

Now why, pray tell, was he discussing how to bring down ‘Darwinism’? Could it be because…

I’m a right wing fundamentalist who takes Genesis literally and wants to see its 6-day creation as mandatory teaching in the public school science curriculum

Now aren’t creationists suppose to be fond of pointing out tautologies? Where are they to catch this one?

I wished they’d have skipped the thumbs and big toes. Then we’d have a hexadecimal system - much more convenient.

A few cultures have used base eight systems. The Chumash Indians of California were an example. Then the Spanish missionaries came along, and assumed that counting in eights was demonic. They missionaries sort of beat it out of them, and made them switch to counting like the Spanish.

About this Entry

This page contains a single entry by PvM published on December 26, 2005 1:17 PM.

Dressing Up ID Is No Substitute for Real Science was the previous entry in this blog.

Reactions to Kitzmiller decision continue is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Categories

Archives

Author Archives

Powered by Movable Type 4.361

Site Meter