See if this provides a basis for the information sought out in IDT:
I know it took longer than a day but here it is. I was debating on not even writing a review, however since you insist on relying on "Complex Specified Information" i figured I'd debunk it.My review of “Intelligent Design as a Theory of information” by William A. Dembski 1998Debunking “Complex Specified Information” and “Intelligent Design”.
With this article, I will show why “Complex Specified Information” and “Intelligent Design” are in fact pseudoscience. Note that the prefix pseudo- is derived from Greek and means false, fraudulent or pretending to be something it is not. Now the obvious question one might ask is how can we identify pseudoscience? I have found the easiest method is to note how words are defined and insure that the same definition is adhered to throughout the entire process of using the scientific method. If the same definition is in fact adhered to, then the subject matter cannot easily be identified as pseudoscience. If however, multiple definitions are used for the same word then we have an easy indication that the subject matter is pseudoscience.
The next question that one might ask is, why is definition adherence so important? The reason is that if someone proposes a definition in science which is not yet accepted, then they should show why the proposed definition is acceptable. In the process of showing why a proposed definition is acceptable, the same proposed definition must be used in the process of showing why it is acceptable, otherwise the proposed definition cannot and should not be accepted.
Pseudo-scientists will sometimes (if not typically) propose a definition which has not been accepted and in the process of showing why the proposed definition should be accepted will instead use an already accepted alternate definition. The goal of the pseudo-scientist is to trick the intended audience into believing that the proposed definition should be accepted when all that was actually accomplished is showing why an already accepted alternate definition Is accepted.
For reference - http://en.wikipedia.org/wiki/PseudoscienceWARNING:
This is a rather long read. Please forgive me. Hopefully I made it as easy and interesting to read as possible.Information
Dembski defines Information as “the actualization of one possibility to the exclusion of others”
. This definition appears to be more in line with “an event having occurred”
within probability theory, and not information (http://en.wikipedia.org/wiki/Probability
). Probabilities are given a value between 0 (0%) and 1 (100%) to indicate the likelihood of occurrence of an event. The actualization of the probability is in fact stating that the “event has occurred”
. The way in which Dembski defines “information” is fairly useless as any event which occurs is by his definition, information. A lightning strike is, according to his definition, information. The Earth continuing to orbit the Sun is, according to his definition, information.
I say the definition that Dembski uses is fairly useless because it is a shallow method of disambiguation in regards to the word “information”, one which is not used within Information Theory. Information, within Information Theory, is defined as “a sequence of symbols that can be interpreted as a message”
). This definition of information would be, contrary to what Dembski asserts, used by Information Theory which was developed by Claude E. Shannon as the transmission of signals across a communication channel.
In reality, “Information” as used in Information Theory, is caused by an intelligent agent and we know who the intelligent agent is. The intelligent agent behind “a sequence of symbols that can be interpreted as a message”
is human beings. By redefining the word “information” to be an event which has occurred, Dembski associates something we know to be caused by an intelligent agent with all events that occur which will cause the reader to naturally assume that events that occur have to also be caused by an intelligent agent.Complex Information
Here we see Dembski attempt to show how to measure information (an event having occurred) in an effort to make his definition useful, which is basically trying to prove my initial conclusion about his definition being useless as incorrect. Dembski states “In measuring information it is not enough to count the number of possibilities that were excluded, and offer this number as the relevant measure of information”
, which seems to be a statement that we can’t simply calculate the probability of “an event having occurred” as relevant measure of information, so Dembski continues by stating “the problem is that a simple enumeration of excluded possibilities tells us nothing about how those possibilities were individuated in the first place.”
So apparently in order to measure “the actualization of one possibility to the exclusion of others”
, we need to individuate the possibilities.
It is true that certain events have a higher probability of occurring than other events. Dembski claims that events with lower probabilities contain more information than events with higher probabilities. This claim by Dembski is also true, for instance learning that someone rolled a 2 after rolling a 6 sided die would be the acquisition of more information than learning that a fair coin flip landed tails, however he chooses an incorrect method of demonstrating his claim.
The example provided by Dembski is being dealt a royal flush as opposed to everything else, when being dealt 5 cards out of a 52 card deck. A royal flush however is individuated by humans, because humans invented the game of p oker. In reality, a royal flush still has the same probability of occurring as any other random 5 cards being dealt, for instance a 2 of hearts, 5 of diamonds, 6 of clubs, jack of diamonds and king of spades, and thus has the same amount of information. The only reason a royal flush would contain more information is if you consider that you’re playing p oker instead of just being dealt 5 random cards. Dembski doesn’t explain how the fact that two sequences of 5 cards (one “random” and another a “royal flush”) with the same probability still somehow allows us to infer that one was designed but the other couldn’t have been designed.
Notice how Dembski returns to a traditional definition of information in this section. The traditional definition of information being “a sequence of symbols that can be interpreted as a message”
. A 6 sided die has 6 symbols, if you roll it 6 times the sequence of symbols can be interpreted as a message. A coin has 2 symbols, if you flip a fair coin 10 times the sequence of symbols can be interpreted as a message. A 52 card deck has 52 symbols, if you are dealt 5 cards the sequence of symbols can be interpreted as a message. Dembski has abandoned his definition of information which was “the actualization of one possibility to the exclusion of others”
Things really start getting hysterical when Dembski attempts to justify his assertion that the obvious way to transform probabilities is with a negative logarithm by stating that the most convenient way for communication theorists to measure information is in number of bits transmitted across a communication channel. I guess Dembski forgot that he himself stated that “The fundamental intuition underlying information is not, as is sometimes thought, the transmission of signals across a communication channel”
Eventually, Dembski claims that information is a complexity-theoretic notion, and I am starting to feel like the word “information” is being abused worse than a 2 dollar crack whore. Yes, somehow with all his mathematics, Dembski has redefined “information” to be “complexity”. Actually, all Dembski has done with his mathematics in this section is define mutual information (http://en.wikipedia.org/wiki/Mutual_information
) and conditional entropy (http://en.wikipedia.org/wiki/Conditional_entropy
Dembski does define “complexity of information” when he states “Given an event A of probability P(A), I(A) = -log2P(A) measures the number of bits associated with the probability P(A). We therefore
<snip> say that the complexity of information increases as I(A) increases (or, correspondingly, as P(A) decreases)”
. This however is not the way in which complexity is defined in Information Theory (http://en.wikipedia.org/wiki/Algorithmic_complexity_theory
). Granted there are many ways to define complexity, why Dembski chooses to not use the common definition of complexity in Information Theory when discussing “Intelligent Design as a Theory of Information” without showing why his definition is acceptable is intentionally deceptive.
As Wesley Elsberry and Jeffrey Shallit note in their article “Information Theory, Evolutionary Computation, and Dembski’s “Complex Specified Information”” (http://www.talkreason.org/articles/eandsdembski.pdf
) – “It is important to note that Dembski’s somewhat idiosyncratic definition of “complexity” is often at odds with the standard definition as used by algorithmic information theorists. For Dembski the string – 111111111111111111111101111111111111111111 – if drawn uniformly at random from the space of all length-41 strings, has probability 2-41 and hence is “complex” (at least with respect to a “local probability bound”), whereas for the algorithmic information theorist, such a string is not complex because it has a very short description.”
What Dembski is doing is quite simply saying that the occurrence of an event with low probability is complex information, thus low probability is complex. Of course, Dembski also acknowledges that more information also makes something complex. I do not agree with Dembski in regards to low probability also being an indication of complexity, as he has failed to show why this is true. In fact, Dembski has offered no examples at all showing why low probability infers complexity, which is contrary to accepted definitions of the word complexity in many other fields of science, including Information Theory.Complex Specified Information
In this section, Dembski defines “Specified Information” as independently given patterned information. “Unspecified Information” is pretty much everything else that isn’t a pattern or a “fabrication”
(which is just a post hoc pattern read off already existing information).
It is interesting that Dembski quotes Richard Dawkins as stating “Complicated things have some quality, specifiable in advance, that is highly unlikely to have been acquired by random chance alone. In the case of living things, the quality that is specified in advance is . . . the ability to propagate genes in reproduction.”
Note: I didn't put the ". . ." in the quote, Dembski did. I have to ask, did Dawkins give Dembski the idea for “Complex Specified Information”? It would seem so. Of course, it is obvious that Dembski is quote mining Dawkins. I suspect that Dembski has taken almost all of the individuals he has quoted in his article out of context. In the case of the quoted Dawkins statement, I know Dawkins is referencing the fact that evolution is not random. Dawkins is not talking about complex specified information.
So we’re clear, according to Dembski, “Complex Specified Information” is an independently given patterned occurrence of an event with low probability. Of course, humans or some known cause are always the intelligent agent behind independent patterns. I am willing to accept this definition of “Specified” provided by Dembski. I do not however accept his definitions for “Complex” and “Information”.
One thing is for certain, CSI can apparently be used to prove that humans exist. Dembski demonstrates this by stating that the 16-digit number on a VISA card is an example of CSI. He states “the complexity of this number ensures that a would-be-thief cannot randomly pick a number and have it turn out to be a valid VISA card number”
, however he abandons his previous definition of the word complexity and uses the traditional definition used in Information Theory.
In algorithmic information theory, the complexity of something is measured by the number of resources needed (such as bits) to specify or describe it. It is the fact that a 16-digit number on a VISA card cannot be described any easier than the number itself is what makes it complex. Probability has nothing to do with making a 16-digit number on a VISA card complex. The same is true of his other examples such as a phone number, the numbers on bills, credit slips and purchase orders. No, Mr. Dembski, CSI is not what makes the world go round. CSI did not motivate the greedy Michael Douglas character in the movie Wall Street
to lie, cheat and steal. The total and absolute control of CSI was not the objective of the monomaniacal Ben Kingsley character in the movie Sneakers
. CSI is not an artifact of interest in any techno-thrillers. CSI does not captivate anyone except for the gullible.Intelligent Design
This section is useless as all Dembski does is attempt to show how CSI can indicate a human Intelligent Agent(designer) or other various animal Intelligent Agents(designers). Low probability and the occurrence of an event have nothing to do with the indication however as the traditional accepted definitions of “Complex” and “Information” are far better indicators of human or known causes. Is CSI found outside of human or other known causes? No it isn’t, and Dembski doesn’t offer any examples in this article (although I know he does in his books, which has been debunked by others and myself already).
Still, even if Dembski was able to prove that CSI exists without human or other known causes, and the conclusion was that some supernatural being was the “Intelligent Designer”, we would be forced to ask, what designed the designer? Subsequently, what designed the designer of the designer? Etc. Etc. This ultimately makes ID self-defeating.
CSI, as defined by Dembski, is not a reliable indicator of design. If I myself were to define CSI - according to acceptable definitions - to make it a reliable indicator of design - I would define CSI as an independently given patterned sequence of symbols that can be interpreted as a message which is complex (as defined by algorithmic complexity theory). Of course, Dembski has no interest in using my definition of CSI, he wants to use low probability of events that have occurred. Proving that an “Intelligent Designer” exists with unproven definitions (simply because certain improbable events with an independent pattern have occurred) is incredibly easy to do.
For example, according to the definition of CSI proposed by Dembski, my finger prints exhibit CSI. My finger prints are an independent pattern, the probability of my exact finger prints existing is unimaginably low and my finger prints exist therefore they are an event which has occurred. I guess that means my finger prints were intelligently designed. My definition of CSI would not prove an “Intelligent Designer” as the fingerprint itself is the symbol however it cannot be interpreted as a message and the symbol isn’t complex as it can simply be described as my fingerprint. For reference: http://en.wikipedia.org/wiki/Fingerprint
Another example would be a snowflake, which according to the definition of CSI proposed by Dembski, exhibits CSI. My definition of CSI however does not show a snowflake to exhibit CSI. For reference: http://en.wikipedia.org/wiki/Snowflake
Notice how Dembski uses the words “general” or “generally” eight times in this section alone as opposed to 5 times in every other section combined. I think even Dembski knows his definition of “Complex Specified Information” is pseudoscientific. The reason Dembski uses the qualifiers “general” and “generally” is so he or anyone can selectively choose what exhibits CSI based on his incoherent definition of CSI. All you have to do is sub in the accepted definitions or sub in the unaccepted definitions proposed by Dembski.The Law of Conservation of Information
Of course in order for Dembski to come up with his new “Law of Conservation of Information”, he does exactly what can be expected when multiple distinct definitions are used for the same word. Dembski is selectively subbing in accepted definitions for the words he’s defined in order to show what cannot exhibit CSI and is subbing in his unaccepted definitions to claim what does exhibit CSI.
The fact is, there is no “Law of Conservation of Information” at least not in how Dembski would define the word information.
The only law regarding information that I am aware of is the one that state that information cannot be destroyed. In that law, information is defined as physical information. For reference: http://en.wikipedia.org/wiki/Quantum_information
- http://en.wikipedia.org/wiki/Physical_information My Conclusions
Dembski never actually supports his proposed definitions of the words contained within “Complex Specified Information” and instead uses acceptable definitions to try and distract the audience (or reader) from this fact. “Complex Specified Information” is garbage pseudoscience intended to prove “Intelligent Design” which is also – by extension – pseudoscience.
Neither “Complex Specified Information” or “Intelligent Design” can make any predictions, are not falsifiable, cannot provide a viable hypothesis and thus fail to use the scientific method. All of these things make both “Complex Specified Information” and “Intelligent Design” not science.
If ID proponents can show how ID can be falsified, provide acceptable definitions, make meaningful predictions and actually use the scientific method, then perhaps it can be considered science and be taken seriously by the scientific community.
Ultimately, “Intelligent Design” should not, without any doubt, ever be taught to anyone, especially school students. At least, not until it can prove to actually be scientific.
If you would like to read about how science actually works, you might as well start with one of my favorite topics - http://en.wikipedia.org/wiki/General_relativity