posted 21. September 2004 15:16                   

quote:

---

13 Conclusions

We have argued that Dembski's justification for "intelligent design" is flawed in many respects. His concepts of complexity and information are either orthogonal or opposite to the use of these terms in the literature. His concept of specification is ill-defined. Dembski's use of the term "complex specified information" is inconsistent, and his proof of the "Law of Conservation of Information" is flawed. Finally, his claims about the limitations of evolutionary algorithms are incorrect. We conclude that there is no reason to accept his claims. Finally, we issue several challenges to those who would continue to pursue intelligent design" as a research paradigm.

---

quote:

---

Rick Sternberg on Process Structuralism

They [process structuralists] also allow themselves to wonder about the cause of the amazing repetition of forms across the biological world rather than being forced by prior commitments to accept a major neo-Darwinian epicycle known as "convergent evolution."

---

quote:

---

Do you believe genetically engineered products evidence CSI by Dembski's definition?

---

quote:

---

I have no need to believe that anything evidences CSI by Dembski's definition. The reason that I need not believe any such thing is that there has never been a successful application of Dembski's EF/DI via the GCEA meeting or exceeding the "universal small probability" of any event whatsoever. If you had a citation of such a published successful, fully worked-out calculation, I'm sure that you would share that with us.

Until then, it's all just blowin' smoke.

---

posted 21. September 2004 20:18                   

quote:

---

Get a lab partner.

Take 2 small boxes (like shoe boxes).
Have the lab partner stand a distance away and not look while you do the following:

Place a configuration of coins in one (like stacked or all heads, or in a pattern), and take another boxes with coins and shake it up.

Present the boxes to someone honestly interested in the experiment, and see if they can detect which box evidences design with respect to the coin configuration.

Then let them do the same for you! They will try to make one box look designed, and let the other box be random. I can almost guarantee, if both participants are honest and determined to make the experiment work, you will have a 100% success rate of detection!

Make any variation of this experiment you want and you will see you can detect ID of even things you've never seen before!

Clearly the box and the coins are designed. The box is one layer of design and the existence of the coins another. We are also aware of an intelligent agency capable of ordering the coins. What the experiment demonstrated was that a layer of design over and above the design of the box and the coins was detectable. We could detect design in the configuration of the coins.

---

quote:

---

If you were presented with a 700-mer polymer made of alternating D-amino and L-amino acids with all 20 types represented and the pattern of chirality was something like:

L L L L D D D D L L L L ....

would you presume it was synthetic given what you know of chemistry?

---

quote:

---

Yes. Natural processes only metabolism L-amino acids, any polymer with D-amino acids must be synthetic.

---

quote:

---

Intelligent causes can do things that unintelligent causes cannot and can make their actions evident. When for whatever reason an intelligent cause fails to make its actions evident, we may miss it. But when an intelligent cause succeeds in making its actions evident, we take notice. This is why false negatives do not invalidate the explanatory filter. The explanatory filter is fully capable of detecting intelligent causes intent on making their presence evident. Masters of stealth intent on concealing their actions may successfully evade the explanatory filter. But masters of self-promotion intent on making sure their intellectual property gets properly attributed find in the explanatory filter a ready friend.
......
The explanatory filter is a net. Things that are desgned will occasionally slip past the net..... This problem cannot be fixed .

Nevertheless we want to be very sure that whatever the net does catch includes only what we intend it to catch, to wit, things that are designed. Only things that are designed had better end up in the net.

---

posted 22. September 2004 12:16                   

quote:

---

Complex Specified Information :

The coincidence of conceptual and physical information where the conceptual information is both identifiable independently of the physical information and also complex.

---

quote:

---

Example:

I have an idea for a novel and the words are in my head before I manage to get to my computer.

I type pages of that novel into my computer.

---

First of all, there can be no doubt of design here, but this point is crucial. CSI is absolutely known to exist in these cases because we were the ones doing the designing!

OK where is the CSI in this illustration?

The words in my head are the conceptual information.

The charaters residing in my computer are the physical information. The coincidence of the two kinds of information is CSI.

This example is extensible to what I call the "Blue Print/ Artifact " metaphor within classes of CSI.

For example, if we genetically engineer something like Bio-Reporter bacteria, we have a BluePrint (the design of such a creature) and then the Aritifact (the actualized, engineered bacteria). The blue print is the conceptual information, and the engineered bacteria conveys the physical information. The coincidence of the conceptual and physical information is CSI.

The question arises how do we identify designs where we don't have the blue prints in hand for??? Elsberry and Shallit 2003 actually inspired a solution with their SAI concept. The coin in the box and some of the designed chemical examples are recognizable as designed objects because they exhibit SAI.

Symbolic information from these CSI-SAI objects exhibit algorithmically compressible phenomenon within Chaitin-Kolmogorov-Solomoff infomation theory, but not just any kind of algorithmic compression (more later on this important topic).

I will call such objects with this particlar kind of CSI-SAI as objects falling under the "order that defied entropy" metaphor, or the "artifacts of maxwell's demon" metaphor.

But I should emphasize, CSI-SAI objects are only one class of objects within the broad spectrum of CSI objects, they are however, a very important class of CSI objects.

For completeness from No Free Lunch page 139:

quote:

---

Conceptual Information: Intelligent agent S identifies a pattern and thereby conceptually reduces the reference class of possibilities

---

quote:

---

Physical Information : Event E occurs and thereby physically reduces the reference class of possibilites.

---

quote:

---

The reference class of possibilities Omega represents the collection of possible events identified by an intelligent agent S.

---

posted 22. September 2004 12:18                   

Consider this example.



Consider a 1000-bin "pill box" (you'll have to imagine it since I only have pictures of those with 7 bins each) and coins in each bin.

For 1000 coins we have 2^1000 possible configurations with respect to heads tails.

I will create an "out of the air specification" independent specification. There are many possible specifications, but let me make one that has a human idiosyncracy.

My specification is: let the Heads/Tails configuration of the first 500 coins be replicated by the last 500 coins.

How many bits does my specification have????

Upon thinking, we realize there are only 2^500 such coin patterns out of the 2^1000 possibilities.

The probability of any such configuration being actualized is

2^500 / 2^1000 = 2^-500

Information content = -log2 (2^-500) = 500 bits of information

(check my figures anyone, thanks)

Using Dembski's definitions, in this illustration:

Omega = the space of all possible coin configurations of those 1000 coins

T = conceptual information = configurations where the Heads/Tails pattern of the first 500 coins is replicated by the last 500 coins

E = the phyisical information = the information conveyed by the coins actually in the 1000-bin pill box.

If E meets the specification of T and E the first 500 bits of E is then the coincidence of the information between T and E is CSI.

Clearly the 1000-bin "pill box" and the coins and the fact the coins in the "pill box" are clear evidence of design. Those are, if you will two existing layers of design that are beyond dispute.

If however one finds a configuration of coins as described above, one presumes another layer of design exists.

Recall again how two parties were able to infer design with the shoe box experiment described earlier. This was a formal examination of such an ID detection method.

This "pill box" specification which I offered describes artifacts which also exhibit SAI.

Realize that I was able to describe 2^500 possible coin configurations succinctly. If I were to attempt to list all of the possible objects in explicit detail I could not do so.

To the formalists out there, one might argue that my specification of the coin strings describe algorithmically compressible phenomenon therefore they are not formally speaking K-complex phenomenon. I don't believe Bill Demski argues that bit streams which evidence CSI must be K-complex.

quote:

---

from Debmski's Design Revolution page 84:

For something to exhibit specified complexity it must have low specificational complexity ( as with the sequence HHHHHHHHHH, consisting of ten heads in a row) but high probabilistic complexity (i.e., its probability must be small). It's this combination of low specificational complexity (a pattern easy to describe in relatively short order) and high probabilistic complexity (something highly unlikely) that makes specified complexity such an effective triangulator of intelligence.

---

So, if Elsberry and Shallit 2003 say that Dembski's usage is "orthogonal" to current literature, I won't speak for that. The word complexity in CSI refers to the improbability of the event. If the complexity theorist consider the usage of the word complexity in the acronym CSI "orthogonal", orthogonal it is. (but the concept still works).

In the formal sense "1000 coins" heads is from a symbolic standpoint not K-complex but I believe the coincidence of "1000 coins heads" with my above specification would evidence CSI. If it makes one feel better we can tighten the specification so that the first 500 coins are K-complex, but establishing that operationally would be intractable as K-theorists are well aware.

The prhase "low specificational complexity" needs to be explored. Dembski has now made it explicit that CSI has a degree of "compressibility" in it in terms of representation of the specifications.

In fact, the 1000-coin-string example above of CSI that is patently not K-complex with respect to the Heads/Tail symbols.



PS and errata
some things I said at antievolution.org and arn.org I have amended in this post to conform to Bill Dembski's usage after my correspondence with him. I would also like to acknowledge and thank Wesley Elsberry for the time he has given in dialogue with me so that this thread would be possible. In fact, I share the nice compliment which Bill Dembski gave to Jeffery Shallit in the acknowledgement section of one of his books:

From Design Inference page xv:

quote:

---

Dembski writes:

As for computational complexity theory, I was introduced to it during the academic year 1987-88, a year devoted to cryptography at the computer science department of the University of Chicago. Jeff Shallit, Adi Shamir, and Claus Schnorr were present that year and helped me gain my footing.

---

posted 22. September 2004 14:35                    

quote:

---

The words in my head are the conceptual information. The charaters residing in my computer are the physical information. The coincidence of the two kinds of information is CSI.

---

posted 22. September 2004 16:20                   

quote:

---

Things that are desgned will occasionally slip past the net..... This problem cannot be fixed.

---

quote:

---

In your metaphor, are you saying the "blueprint" is in your head?

---

quote:

---

How do you propose to establish that the conceptual information is identifiable independently from the physical information, i.e., meet the detachability criterion?

---

First let's take trivial, and uninteresting examples.

In the case of a password that you create, it's easy to establish detachablity. When you are setting up your computer account and you are prompted to supply a password, the event of you supplying the password to the computer will not happen until you actual type it in. Before you type it in, the specification is still in your Mind and detached from the actualized event. This is clearly trivial, but I hope it begins to clarify what CSI actually is.

In the case of engineering a space ship, the manufacturing process has not happened yet and the blue prints are still just blueprints. In the case of building a space ship the blue prints (designs) are the conceptual information, and the characteristics of the eventually manufactured space ship is the physical information. The reason the blue print is consider "detached" (not post-dictive) is that the initial specification (blue print) was not affected by the actual event which the specification helped actualize (manufacture).

These trivial and unintersting cases are important in that it at least gives a clue what "detachable" really means. What has happened is that CSI has been used so much in the origins debate, that fundamental and easy aspects about CSI are lost in the discussion.

Now, the question of the day is, how do we establish detachable specifications for things that have already happened? How do we establish what we think the "blueprint" was? It's the whole problem avoiding post-diction, or after-the-fact creation of specifications.

So how do we establish specifications after the event already took place??? It is that problem which Elsberry and Shallit 2003 actually helped solve (albeit unwittingly, and I'm sure they might protest at me crediting them for doing so ). They identified the class of CSI specifications that exhibit SAI! SAI I suspect is Shallit's term, but Dembski refers to the idea of SAI (even before the word was coined) in No Free Lunch page 144:

quote:

---

It is CSI that within the Chaitin-Kolmogorov-Solomoff theory of algorithmic information identifies the highly compressible, nonrandom strings of digits.

---

quote:

---

Bill must mean something else. I can't say I'm enamored with his terms here. Perhaps the conceptual information in your head is there as a result of reading the novel. How will you convince me otherwise? I certainly know that I have read far more novels than I have written.

---

quote:

---

Perhaps the conceptual information in your head is there as a result of reading the novel.

---

When you exercise your will to plagarize (even unconciously), the conceptual information still originated from someone's mind, if not necessarily all from yours. The ownership of the specification is admitedly in question, but it is still a specification. Oddly enough, you have touched on the important issue of displacement.

[ 08. July 2005, 12:45: Message edited by: Salvador T. Cordova ]

posted 22. September 2004 16:37                   

quote:

---

Continuing in the same vein, Dembski repeats his often-stated thesis that what he calls "specified complexity" is a necessary indicator of design.
.....

Indeed, consider an example discussed several times before (Perakh 2001): Imagine a pile of pebbles found on a river shore. Usually each of them has an irregular shape, its color varying over its surface, and often its density also varying over its volume. There are no two pebbles which are identical in shape, color, and density distribution. I guess even Dembski would not argue that the irregular shape, color, and density distribution of a particular pebble resulted from intelligent design, regardless of how complex these shapes and distributions may happen to be. Each pebble formed by chance. Now, what if among the pebbles we find one that has a perfectly spherical shape, with an ideally uniform distribution of color and density? Not too many people would deny that this piece in all likelihood is a product of design. However, it is much simpler than any other pebble, if, of course, complexity is defined in a logically consistent manner rather than in Dembski’s idiosyncratic way. A logically consistent definition of complexity is given, for example, in the algorithmic theory of randomness-probability-complexity (and is often referred to as Kolmogorov complexity). The Kolmogorov complexity of a perfectly spherical piece of stone is much lower than it is for any other pebble having irregular shape and non-uniform distribution of density and color. Indeed, to describe the perfectly spherical piece one needs a very simple program (or algorithm), actually limited to just one number for the sphere’s diameter, one number for density, and a brief indication of color. For a piece of irregular shape, the program necessarily must be much longer, as it requires many numbers to reproduce the complex shape and the distributions of density and of color. This is a very simple example of the fallacy of Dembski’s thesis according to which design is indicated by "specified complexity." Actually, in this example (as well as in an endless number of other situations) it is simplicity which seems to point to design while complexity seems to indicate chance as the antecedent cause of the item’s characteristics.

---

quote:

---

from Debmski's Design Revolution page 84:

For something to exhibit specified complexity it must have low specificational complexity ( as with the sequence HHHHHHHHHH, consisting of ten heads in a row) but high probabilistic complexity (i.e., its probability must be small). It's this combination of low specificational complexity (a pattern easy to describe in relatively short order) and high probabilistic complexity (something highly unlikely) that makes specified complexity such an effective triangulator of intelligence.

---

Another word for "low specificational complexity" is SIMPLICITY!

CSI-SAI are cases where simplicity is the evidence of design. Simplicity in this case is "order that defied entropy". But simplicity is not a mandatory requirement for design, simplicity in certain cases tends to be the easiest to identify design, at least in human affairs.

quote:

---

Mark Perakh wrote:

complexity is defined in a logically consistent manner rather than in Dembski’s idiosyncratic way.

---

posted 26. September 2004 22:42                   

quote:

---

An alternate view is that if specified complexity detects anything at all, it detects the output of simple computational processes. This is consonant with Dembski's claim "It is CSI
that within the Chaitin-Kolmogorov-Solomonoff theory of algorithmic information identifies
the highly compressible, nonrandom strings of digits" [19, p. 144].

---

quote:

---

Salvador wrote:
I invite Wesley care to quantify thephrase "simple computational process"?

---

quote:

---

That would be the appendix detailing "Specified Anti-Information".

---

quote:

---

Dembski's inference of design is then undermined by the recent realization that there are many naturally-occurring tools available to build simple computational processes. To mention just four, consider the recent work on quantum computation [42], DNA computation [47], chemical computing [55, 89, 74], and molecular self assembly [79]. Furthermore, it is now known that even very simple computational models, such as Conway's game of Life [3], Langton's ant [26], and
sand piles [33] are universal, and hence compute anything that is computable. Finally, in the cellular automaton model, relatively simple replicators are possible [5].

---

(bolding mine)

I found the bolded portion to be an astonishing claim. I realized that the word "naturally occurring" was being equivocated here. I believe the equivocation hints that such "naturally occuring computations" are the product of undirected natural forces.

Having worked with nano-molecular technology I've not seen any of those examples naturally arise! Sure once the intelligently designed hardware and software are in place, then things "naturally arise", but it's like saying a cake naturally arises after the recipe (design) is followed.

An internet search will uncover that DNA computation is a designed activity. Dr. Aldeman's work on DNA computing has been highly commended to the group I did scant work for. DNA computers are anything but naturally arising.

Further Quantum Computers are anything but naturally arising, sure there is a thing called "natural computing series", but look at what that book actually explores:

Quantum Computing : Natural Computing Series

If one looks at the description of the book, "natural computing series" has nothing to do with undirected nature creating computational processes any more than the phrase "natural numbers" imply that undirected nature can infuse coin strings with profound mathematical "natural numbers".

Quantum computers have to be intelligently designed to work, and they are anything but natural and simple. Had such machines naturally arisen, Intel would be out of business. I therefore found it astonishing Wesley would appeal to such systems. He's welcome to correct me if I'm misinterpreting what he is saying, but it just seems grossly incorrect.

Molecular self-assembly is a hot topic in nano-technology, and self-assembly is a pre-programmed design approach for nano-molecular machines to self-assemble. Those cannot therefore be used to remotely suggest undirected natural forces cause nano-molecular machines to "self-assemble" without intelligent design.

Similar equivocations with Langton Ant, game of life, and chemical computing. Readers of this thread are invited to elaborate on these supposed examples of "natural" computation.

Salvador

[ 30. September 2004, 00:00: Message edited by: Salvador T. Cordova ]

posted 04. December 2004 11:11                   

quote:

---

Dembski in No Free Lunch

Complex Specified Information :

The coincidence of conceptual and physical information where the conceptual information is both identifiable independently of the physical information and also complex.

---

Now looking at Elsberry and Shallit 2003

notice the following key phrases were missing:

1. "physical information"
2. "conceptual information"
3. "coincidence of"



The crucial phrase is: CSI is "coincidence of conceptual and physical information". To write a paper to refute CSI and not include the most central definition of CSI is inexecusable.