Evidence Contrary to Evolution - Information Theory

Absolutely everyone agrees that biological systems give the appearance of being designed. In fact, there's agreement that 'design' requires a designer with intelligence. The disagreement comes in when we discuss the question of whether or not biological systems really are designed.

Wouldn't it be nice if there was an objective scientific measure of what constitutes 'design'? Well, there is. It's found in the discipline of Information Theory. William Dembski is perhaps the most well-known proponent of using information theory to judge whether biological systems are the product of an intelligence (it's universally assumed that information content only comes from intelligence).

The genetic code itself is ideal for the study of informational content. The code is made up of a long sequence of 4 repeating nucleotides very much like a short alphabet. In information theory, there are two attributes that both must be contained by some pattern to be considered to contain information. These 2 attributes are complexity and specificity. Here is an example of a string of highly complex symbols:

lkaklsfdoiuwernlkjnahiasvfoias

Notice that this string of data is complex in that the number of different combinations that could occur here (using a 26 character alphabet and 30 characters) is huge. So this fits the complexity test. But what about specificity? Is there anything specified about the seemingly random string of letters? No. A bunch of letters layed out this way randomly does not contain information.

Here is something that is very specified, yet not complex:

aaaaaaaaaaaaaaaaaaaaaaaaaaaaaa

There is no reason to assume information here because, although highly specified there is nothing complex about a recurring string of characters.

This on the other hand:

once upon a time there lived a

Is both complex (huge number of different possible arrangements) and specified (the arrangement is far from random). Now in this case, the specificity is defined by English grammar. If you were to see an apparently random string that was the result of an encryption algorithm, how could it be distinguished from random characters? Well, even an encrypted string of data can be detected to be informational via information theory. There are algorithms that can detect specificity even in the absence of understanding the rules of that particular specification. For instance, when an encrypted transmission is received by the NSA, they can first analyze it to determine if the message is actually meaningful before embarking on the task of trying to break the code.

Now, when applied to the genetic code that drives all living things can we say this is information? Does it contain the property of specified complexity? The answer is yes. There is definitely complexity. No one denies that the genetic code is huge in terms of the number of different possible codes available. Each living organism since the beginning of time (forgetting Dolly and other cloning experiments) has had a unique DNA signature.

The real question is whether the DNA code is specified. Well, that's hard to deny since we know that there is really a very small subset of possible DNA codes that would result in a living creature. For example, take the DNA of an egg and randomly scramble it. What are the odds you could fertilize it and get a living chicken (or living anything)? Very nearly zero.

Since DNA can be proven to be highly complex and highly specified, the debate is really over the premise of information theory. Can 'information' arise naturally? The burden of proof lies with the evolutionists because there is no example known to man. It would be committing a logical fallacy to look at life on Earth as an example since it would be begging the question.

According to William Dembski, this problem is intractable in that future scientific knowledge cannot change the outcome of the answer. He says it's an ontological rather than an epistimological issue.