Re: A Response to Dr. Dawkins' "Information Challenge" (Part 1): Specified Complexity Is the Measure of Biological Complexity

On Sep 27, 11:25 am, NKTB <north_korean_tourist_bo...@xxxxxxxxx>
wrote:
http://www.evolutionnews.org/2007/09/a_response_to_richard_dawkins.html

A Response to Dr. Dawkins' "Information Challenge" (Part 1): Specified
Complexity Is the Measure of Biological Complexity

Last week I posted a link to a YouTube video where Richard Dawkins was
asked to explain the origin of genetic information, according to
Darwinism. I also posted a link to Dawkins' rebuttal to the video,
where he purports to explain the origin of genetic information
according to Darwinian evolution. The question posed to Dawkins was,
"Can you give an example of a genetic mutation or evolutionary process
that can be seen to increase the information in the genome?" Dawkins
famously commented that the question was "the kind of question only a
creationist would ask . . ." Dawkins writes, "In my anger I refused to
discuss the question further, and told them to stop the camera."
Dawkins' highly emotional response calls into question whether he is
capable of addressing this issue objectively. This will be the first
installation of a 3-part response assessing Dawkins' answer to "The
Information Challenge."

What Type of "Information" Is Relevant Here?

This is the important point. You want to distinguish between
"information" and "information that has meaning" to somebody. But
absent knowledge about the way that the information was manufactured,
you cannot determine how it was manufactured merely by claiming that
it is "information that has meaning".

Dawkins writes, "First you first have to explain the technical meaning
of 'information'." While that sounds reasonable, Dawkins pulls a bait-
and-switch and defines information as "Shannon information"-a
formulation of "information" that applies to signal transmission and
does not account for the type of specified complexity found in
biology.

Although there are people who post here who are far better at
information science than I am (and I am sure they will correct my
mistakes), there are two ways to measure *amount* of information in a
string of symbols without having any knowledge of manufacture.
Neither deals with the 'meaning' in a string of symbols.

One measure (and I believe it is Shannon) assumes an initial message
(string of symbols, whether meaningful or not) is susceptible to
degradation during transmission and starts out at the highest possible
level of order. In this measure, one can measure only loss of
information by defining loss of information as amount of difference
from the originally created message. In some ways, this is
"creationist friendly" because, *by definition*, one assumes that the
current message has the maximum amount of information (the intended
message) and *any* change is therefore a loss of information.
Unfortunately for creationists, this does not prevent a changed
message (with subsequent loss of information) from being *more*
meaningful than the original in some contexts. That is *because*
change in meaning has no input into determining whether or not there
has been loss of information.

The second measure of "information" is counter-intuitive. It is the
idea that the compressibility of a sequence is a measure of how much
information it would take to re-create it. In this measure (I believe
it is Kolomogorov), a completely random sequence is less compressible
and thus has more information than sequences that can be described by
simple algorithms.

It is common for Darwinists to define information as "Shannon
information," which is related to calculating the mere unlikelihood of
a sequence of events. Under their definition, a functionless stretch
of genetic junk might have the same amount "information" as a fully
functional gene of the same sequence-length. ID-proponents don't see
this as a useful way of measuring biological information.

It is fine that ID-proponents don't see this as a useful way of
measuring biologicial information. But they do have to tell us *how*
to quantitatively measure the amount of information in whatever method
they choose to use and why they think it can *usefully* distinguish
between designed and non-designed forms of manufacture.

ID-
proponents define information as complex and specified information-DNA
which is finely-tuned to do something.

That doesn't tell us how you measure the amount of information
differently from any other measure. It only tells us that you are
arbitrarily deciding to only measure the amount of information in
sequences that *you* determine are "finely-tuned" [how do you
determine what is 'finely-tuned' when that depends on the
environment?] to do something.

Stephen C. Meyer writes that ID-
theorists use "(CSI) as a synonym for 'specified complexity' to help
distinguish functional biological information from mere Shannon
information--that is, specified complexity from mere complexity." As
the ISCID encyclopedia explains, "Unlike specified complexity, Shannon
information is solely concerned with the improbability or complexity
of a string of characters rather than its patterning or
significance."

Again, I see no difference other than that you are arbitrarily
deciding to only measure the information content in *some* sequences
that have meaning to you. Either you are measuring information or you
are not. You do not get to choose to look only at genetic sequences
that have meaning to you. Organisms reproduce both useful sequences,
harmful sequences, and meaningless (or neutral) sequences.

The Inconvenient Truth for Dawkins: The difference
between the Darwinist and ID definitions of information is equivalent
to the difference between getting 10 consecutive losing hands in a
poker game versus getting 10 consecutive royal flushes. One implicates
design, while the other does not.

Sure. If you only count hands with royal flushes as having meaningful
information, all hands with meaningful information will have royal
flushes. Organisms, OTOH, do not choose to replicate *only*
information that has meaning. And the environment does have an effect
on what information is "meaningful" (meaning having utility to the
organism's reproductive success, which is how "utility" is defined in
living organisms). A gene can be useful (meaningful) in one
environment, and detrimental (worse than useless) in another. So, if
the environment changes, loss-of-Shannon-information (any change from
the original message is a loss of Shannon information) can increase
the "meaningfulness" or "utility" of the sequence. How does CSI
determine the amount of information present when this happens?

It is important to note ID proponents did not invent the notion of
"specified complexity," nor were they the first to observe that
"specified complexity" is the best way to describe biological
information. My first knowledge of the term being used comes from
leading origin of life theorist Leslie Orgel, who used it in 1973 in a
fashion that closely resembles the modern usage by ID proponents:

[L]iving organisms are distinguished by their specified
complexity. Crystals are usually taken as the prototypes of simple,
well-specified structures, because they consist of a very large number
of identical molecules packed together in a uniform way. Lumps of
granite or random mixtures of polymers are examples of structures
which are complex but not specified. The crystals fail to qualify as
living because they lack complexity; the mixtures of polymers fail to
qualify because they lack specificity.

Actually, crystals are informationally *simpler* than granite in that
they are much more compressible into a simple algorithm. But Shannon
information only applies as they grow and is basically a measure of
how perfect a particular crystal is (how well it follows the algorithm
of growth). Shannon information does not apply to granite precisely
because it is not a message that gets transmitted in any sense.

(Leslie E. Orgel, The Origins of Life: Molecules and Natural
Selection," pg.189 (Chapman & Hall: London, 1973).)

Orgel thus captures the fact that specified complexity requires both
order and a specific arrangement of parts or symbols. This matches the
definition given by Dembski, where he defines specified complexity as
an unlikely event that conforms to an independent pattern.

Not really. Dembski is *himself* choosing to look only at those
sequences that he declares have biological "meaning" (utility to the
organism wrt reproductive success in a particular environment). He is
drawing the bull's eye on the barn around only those sequences that
have meaning and *ignoring* the fact that there is no difference in
the amount of change wrt those sequences and the non-meaningful
sequences that surround it. The only thing that differs is that the
*environment* determines that only those changes that increase or
maintain *utility* (measured by relative level of reproductive
success) have the opportunity to become the message of the future.

This
establishes that specified complexity is the appropriate measure of
biological complexity. This point will be important in the next
installment, Part 2, which rebuts the heart of Dawkins' article.

As a final note, Richard Dawkins' article admits that "DNA carries
information in a very computer-like way, and we can measure the
genome's capacity in bits too, if we wish." That's an interesting
analogy, reminiscent of the design overtones of Dawkins concession
elsewhere that "[t]he machine code of the genes is uncannily computer-
like. Apart from differences in jargon, the pages of a molecular
biology journal might be interchanged with those of a computer
engineering journal." (Richard Dawkins, River Out of Eden: A Darwinian
View of Life, pg. 17 (New York: Basic Books, 1995).) Of course,
Dawkins believes that the processes of random mutation and unguided
selection ultimately built "[t]he machine code of the genes" and made
it "uncannily computer-like." But I do not think a scientist is
unjustified in reasoning that in our experience, machine codes and
computers only derive from intelligence.

All analogies to human-made and human-designed inanimate objects or
systems are necessarily faulty when applied to organisms that
reproduce genomes in the absence of any 'designer'. A car does not
reproduce; a cat does.

Regardless, in the next
installment, Part 2, I will assess Dawkins' argument that gene
duplication can increase biological information.


.