Mr Carcer

What exactly do creationist mean when they say that animals can adapt but there's never any increase in information? There have been quite a number of times when I've been told by creationist that what we think of as new traits are actually old traits recombined. A prime example I was recently given was that humans have such wide variety of physical traits because Adam and Eve's original perfect DNA has been rearranged so many times, but that there is no new information because its just a jumble of old information.

Maybe I'm misunderstanding their stance, but isn't that akin to saying Newton's Principia provided us with no new information because it was really just a recombination of the letter that Milton used to write Paradise Lost?

RBH

That's a good question. On any formal definition of "information" of which I'm aware (e.g., Shannon Information, algorithmic information, etc.,) there are well known evolutionary mechanisms that can alter (increase or decrease) the amount of "information" in the genome of both individuals and populations.

Evolution is descent with modification, and so genuine novelty is very rare. Looking across eucaryotes, for example, the basic functional 'modules' -- e.g., those having to do with energy transformations in cellular metabolism -- are pretty much the same across taxa. That's precisely what we'd expect from common descent. Functions that are less basic -- e.g., motility -- are carried out by a variety of structures, but even there we find deep developmental and genetic homologies: descent with modification again. The 'old' traits are not merely "recombined", they're transformed. And often that transformation involves the cooption of the 'old' trait to serve a new function, as forelimbs in vertebrates were coopted to perform a flight function in birds or more radically, how the sesamoid bone in pandas was coopted to function as a sort of thumb. Yeah. That's a nice analogy, though the unit of analysis is on a level closer to words than letters. Most novels draw from pretty much the same universe of words, but very occasionally a neologism gets into that universe. But even there, the neologism uses the same basic components: letters.

RBH

NonHomogenized

Personally, I'd say that both 'words' and 'letters' could be considered to be excellent analogies, here.

"Words" are, in some ways, akin to proteins: There are a lot of them, they're all made up of the same basic subunits, they can be work together to express more than by themselves, and there are a vast number of them. The major flaw in this analogy would be that, unlike in the case of proteins, you can't just combine any number of random letters and get a word that means something (unless, of course, you assign a meaning to that "word").

Words can also be analagous to codons. I like this analogy a little better, since there are a limited number of each. Further, the context is important; A sequence of codons without start and stop codons is meaningless, much like a sentence without any syntax, and a given word in different context can have a very different meaning, much like placing the same amino acid at the same position in two very different proteins.

Letters are more akin to nucleotide bases, although you could also draw a parallel between letters and codons. However, I like the 'base' analogy a bit better, because everyone accepts that the same letter in a different word is still that letter. When dealing with creationists, right up to the level of "protein", they seem willing to acknowledge the same principle, but as soon as we get multiple modified proteins, it often suddenly becomes an issue of "no new information", completely ignoring that those proteins are still made up of the same 20 amino acids, which are coded for by sequences of those same 4 nucleotide bases.

The "no new information" canard seems, to me, to derive directly from an utter lack of understanding of how genetics work. I think Mr Carcer hit the nail on the head with his analogy. Of course, RBH just had to go and extend it to make it doubly apt, but I think the fundamental reason the analogy is useful is the same between both formulations.

Martin B

Why are you asking us? We've been asking creationists this for a long time and still haven't gotten a straight answer. Instead, we get the same terse answer they give us when we ask what a "kind" is: "You know what we mean!"

Not to mention how comparatively insignificant genetic novelty has been in the evolution of a great deal of morphological novelty. It's becoming old news now that the developmental "toolkit" of animals is pretty much the same in all taxa. The variations tend to be in the number of copies of each gene or in how they are deployed during development. It would be an overstatement if I said no novelty was required, but it is remarkable how little is really required to explain the evolution of animal body plans.

Shows how far you can get by "jumbling old information".

uberhobo

I think the letters/word analogy will wind you up in trouble with a creationist beause they could argue that words themselves are meaningless, and that we ascribe meaning to words. There's is nothing intrinsic to the meaning of the word "red" contained within its letters. That's the base difference between the chemicals of life and words. The intrinsic properties of chemicals dictate their actions and interactions. No matter what arrangement of atoms you decide to make, they're going to do something on their own. The same cannot be said for "woeiurpalkdsjfoe."

Or, you could just be lazy and say that the word "ping" intrinsically describes that which it is supposed to represent, as in, "the machine that goes 'ping.'"

Dlx2

Most morphological innovation isn't done via an increase in information, but rather a change in where, when, and how instructions are applied. However, it's obvious that there have indeed been increases in information when we look at gene and genome duplications. Which basically means creationists are more or less lying through their teeth.

Dhaeron

I'd advise you to use the latter example because you could damage your vocal cords trying to pronounce "the machine that goes woeiurpalkdsjfoe".

LambdaCalculator

According to Greg Chaitan, one of the founders of algorithmic information theory (AIT), the basic idea of the theory, absent the esoteric mathematical jargon, is that you measure computer programs by the smallest amount of binary bits used to specify whatever you want your program specify.

Essentially, when used a programming technique, it's a way of minimizing the resources that programs use in Turing machines, usually to optimize performance. In a language like C, Pascal or FORTRAN, achieving algorithmic information is usually easier than in a higher level language like Ruby or Python (the latter two of which use the bytecodes of lower level languages in their interpreters).

I have a hard time understanding why this "designer" posited by the ID theorists is believed to be intelligent. Given that these techniques are known to programmers after less than a century of computer science research, I would expect some awesome cosmic designer to also be familiar with them or something better. If he did we wouldn't see the kind of redundancy we see in the human genome, and a whole lot of base-4 code that does nothing.

Mike Elphick

In a typical creationist discussion, they frequently turn one's questions round to ask a counter question in an attempt to avoid answering. I suggest you try the same technique with genetic information.

Most creationists are reading from a list of anti-evolution statements. These are practically the same wherever you look. Regarding information, they insist that since God's curse, after the Fall of Man, all genetic information is slowly degrading and getting lost.

Ask them to provide evidence that demonstrates continuous degradation and loss of genetic information. You can point them to the vast amount of literature regarding the genomes of bacteria, dogs, cats, chimps and man himself. Allow them to interpret the data and to show you where information is getting lost.