Werner Gitt
From SkepticWiki
Contents |
[edit] Introduction
Werner Gitt is a creationist. His chief contribution to creationism is a set of maunderings about "information theory" which well illustrate the notorious creationist tendency to, figuratively, dress up in lab coats and play at being scientists. A fair sample of his work (if one can call it that) can be found in his paper Information, science and biology.
[edit] Gitt's writings
The essence of Gitt's writing is to state various creationist dogmas and to call them "theorems" of "information theory". Some samples are given below:
- Theorem 4: A code is an absolutely necessary condition for the representation of information.
- Theorem 5: The assignment of the symbol set is based on convention and constitutes a mental process.
- Theorem 9: Only that which contains semantics is information.
- Theorem 10: Each item of information needs, if it is traced back to the beginning of the transmission chain, a mental source.
Gitt writes, with apparent pride in his work:
- The most important empirical principles relating to the concept of information have been defined in the form of theorems.
Now, this sentence is nonsensical to anyone familiar with the vocabulary of science. It is like hearing someone pretending to be knowledgeable about sport saying: "The quarterback did an end-run around the fifth baseman, so the referee awarded a slam-dunk."
Furthermore, every part of it is wrong: these are not theorems of information theory, they are not definitions in information theory, and they have not been proven empirically; as we shall discuss below.
[edit] Are Gitt's claims theorems of information theory?
Mr Gitt's "theorems" are not theorems of information theory. Information theory is a body of mathematics, and the theorems of information theory are the logical consequences of the axioms and definitions of information theory. Mr Gitt has not thus derived any of his so-called "theorems of information theory", nor even pretended to. He has simply written down some of his personal beliefs and called them "theorems". We might as well reply by writing:
- Theorems of information theory
- Theorem #1: Werner Gitt is talking nonsense.
- Theorem #2: All the information in DNA was produced by evolution.
- Theorem #3: I win.
This too is not information theory: like Mr Gitt, we have simply written down our beliefs and called them "theorems" of "information theory". What the heck does that prove?
The reader who would care to see some real information theory can do no better than to read Shannon's foundational paper on the subject, A Mathematical Theory of Communication.
[edit] Are Gitt's claims definitions in information theory?
Nor are Mr Gitt's beliefs definitions of information theory. For information theory already exists, and already has axioms and definitions. Mr Gitt cannot honestly claim his beliefs to be definitions in information theory any more than he can honestly claim them to be definitions in Euclidian geometry. Not only are Gitt's definitions totally different from the definitions in real information theory, they starkly contradict that theory. For example, in Gitt's third "theorem" he writes:
- Since Shannon’s definition of information relates exclusively to the statistical relationship of chains of symbols and completely ignores their semantic aspect, this concept of information is wholly unsuitable.
In real information theory, information is a quantitative concept, and the sequence: "HTTHTHHHTHTHTTHHT" contains the same amount of information whether it was produced randomly by flipping a coin or whether, on the other hand, it was a coded message between two people meaning "meet me at midnight". But Gitt writes: "A sequence of symbols does not represent information if it is based on randomness"; by Gitt's "axioms", the same sequence contains no information in the first place and an (unquantified) amount of information in the second, so that in order to know if we could even apply "information theory" (in Gitt's sense) we would have to know about the history of the sequence.
Gitt is attempting here to skip over the process of actually doing any science by defining information as the product of intelligence. But then we would need to know if the information in DNA was magicked into existence by God or, on the other hand, the product of natural processes, in order to decide whether there is any information in DNA according to the definition of "information" given by Mr Gitt. By analogy, if Mr Gitt was a devout believer in the Norse gods, he might choose to redefine "lightning" as "the effect produced when the thunder-god Thor swings his magic hammer Mjöllnir". But this definition does not answer any substantive question about lightning: it would merely raise the question of whether there is any such thing as "lightning" in Mr Gitt's sense of the word.
[edit] Have any of Gitt's claims been proven empirically?
To this question we may give a resounding "no". If creationists could prove that all the information in DNA was the product of intelligence, then their task would be complete, and they would be in no need of this smokescreen of nonsense about "information theory".
But such proof has not been forthcoming. Nor does Mr Gitt put up even a shred of a pretense of empirical proof: for his paper contains no observations whatsoever that could possibly be relevant to the origin of information in DNA. Instead, he has simply written down his dogma, called it a "theorem", and claimed victory.
As Bertrand Russell once remarked of such intellectual charlatanism: "The method of postulating what we want has many advantages; they are the same as the advantages of theft over honest toil."
Actual observation proves that Gitt is wrong. For we can see information (in the sense in which that term is used in real information theory) being added to the gene pool by natural and unintelligent processes: see our main article on Beneficial Mutations for some biologically interesting examples. Mr Gitt might deny that this is "information" in his sense: indeed, it is not. But in that case we should have to ask again for evidence that there is any "information" (in Gitt's peculiar sense of the word) in any naturally occurring sequence of DNA.
[edit] Consequences of Gitt's deceit
The result of Gitt's writings is that lay creationists go about saying that the science of "information theory" has proved them right. But they are deceived.
In the first place, they should be aware that Gitt's blather is not part of the science of information theory any more than it is part of General Relativity or thermodynamics. Gitt is merely trying to pass off his writings as science by giving it the same name as an existing body of science. But calling it "information theory" doesn't make it so.
In the second place, he has proved nothing: again, calling something a "theorem" doesn't magically mean that you've proved it; for it to be a theorem, one needs actual proof, and merely writing the magic word "theorem" in front of a statement of one's beliefs is not a substitute.
And in the same way, if Mr Gitt took a medal, called it the Nobel Prize in Information Theory, and pinned it on his own chest, that wouldn't make him a Nobel Laureate.
