Undesigned information

Supplementary material

For a few decades, now, a creationist mathematician called William Dembski has been quite vocal about the concept of "irreducible complexity", which in a nutshell is something (some natural object or phenomenon) that is so complex that it could never, ever have developed without the assistance of magic. Needless to say, the proponents of such a concept usually say that their argument is based solely on science and not on, say, their religious beliefs, but by a coincidence that is even more astonishing that the spontaneous appearance of life, all of them happen to be religious. Dembski himself has a degree in divinity, even! (As a lapsed catholic, I have no precise idea what that is. It is not a degree in theology, and in the U.S. I believe that it is given by church-related institutions (and often as a honorary title). Not to make light of it, but to me it's akin to a degree in Pokémon studies; it means that someone has talked a lot about something that doesn't exist).

What a creationist biochemist like Michael Behe sees as irreducible complexity that proves the need for a supernatural engineer, all other scientists see as systems that may look very complex indeed but that don't require magic to be understood. The famous case of the bacterial flagellum is a good example of irreducible complexity that is after all quite reducible. It is a complex biochemical machine that bacteria use to move about, and although it's true that it is unlikely for each piece of this complex apparatus to have evolved independently until *shazam!* all were ready to be assembled to form a fully-functional flagellum, it is far less flabbergasting when we realize that most of these pieces were already present in bacteria, and used for other purposes. A sandwich appears less impressive in its final form when you know that the fridge contained ham and cheese and mayo and that there is bread in the pantry and a tomato on the kitchen counter.

"If you can't dazzle them with data, baffle them with bullshit". In my case, that would be very easy to do if the speaker was a mathematician, because my math-fu is weak. So when a mathematician says things like "information can neither be created nor destroyed", I simply wouldn't know where to start looking if Google didn't exist. Many creationists, who are as bad as I in mathematics and decidedly worse in biology, often refer to that line from Dembski as if his views on information theory were accepted by the scientific community, clearly thinking that they are as sound as Newton's laws of thermodynamics. Equating the succession of nucleotides in a DNA sequence with information, these enthusiasts readily quote their champion: "neither algorithms nor natural laws are capable of producing information."

That DNA contains information, I have no problem with. Obviously, the succession of nucleotides forms a code (it's called the genetic code, after all!) that directs the assembly of proteins, which in turn help living cells function. On top of that, the shorthand way to describe a succession of nucleotides is to drop the phoshoribosyl backbone of DNA and just focus on the first letter of each nucleotide's name: A, C, G or T. To a casual reader, it really looks as if DNA is "written" in some fashion. Like a message in morse code, or a computer line written with 1s and 0s. But that's where the analogy goes too far.

Take the following line from Shakespeare: "There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy". It is clearly information, and this information is carried by a totally arbitrary system of communication: certain signs arranged in a certain order, and the message itself is totally independent of the communication system used. You could write that sentence in Chinese or Korean and the information itself would not change.

Not so with the genetic code. Each nucleotide is more than a symbol; it is a molecule whose very shape will determine its possible interaction with other molecules. In that sense, it is more like a cog in a machine that a letter in a sentence. But even the cog analogy is abusive; a cog in a machine is placed just so, and other cogs are likewise finely positioned; no matter what William Paley might elegantly argue, a watch does not contain a bucketload of useless or broken cogs. The human genome, by contrast, is full of parasitic transposons and of integrated retrovirus remains, not to mention thousands of pseudogenes.

But there is information anyway in our DNA, even if this information is mechanistic in nature more than based in an abstract code. Does the statement "neither algorithms nor natural laws are capable of producing information" apply, then? Well, no, I beg to differ. As the cartoon above illustrates, a mindless flow of water subject to purely natural laws can certainly create information. Looking at a dry river bed, it is clear where the water will flow during the next flash flood. This information is a lot like the genetic code, which instead of directing water in one direction will direct how a transfer RNA will sit next to another during the process of assembling amino acids into proteins. Information can definitely originate from mindless, undirected and clearly not supernatural sources.