We Should Live - Ben Bateman

The Intelligent Design debate is an acquired taste, because it’s initially so bitter. For most, it’s just a battlefield on which to fight the culture war: It’s a way to fight on behalf of oppressed Christianity, or a way to give Christianity a well-deserved poke in the eye. Many participants don’t even seem to understand that there is (or could be) a serious and reasonable discussion of the topic itself, given the noise and raw hatred that the larger battle generates. Observe some of these internet brawls for a while, and you’ll find just beneath the surface chatter about genes and natural selection that each side yearns to scream, “Don’t you see? The other side is just evil!”

Try to leave all that aside. Instead, let’s focus on the substance.

First, let’s define the dispute: The word ‘evolution’ means all sorts of things to different people. For some, it’s about abiogenesis, or how life originally formed out of non-life. For others, it’s about the narrow question of whether humans descended from other primates. Others have difficulty distinguishing evolution from natural selection—or they believe that the other side has that difficulty.

But the proper focus for an interesting intellectual discussion is on how new species emerged. There isn’t much serious debate on the extent to which different species are genetically similar, or on whether natural selection causes some amount of genetic variation within a species. The question is where new species come from, especially species with big new features, like wings, eyes, fingers, etc.

There are two basic views: One holds that it all happens gradually. Natural selection can steadily favor a particular size, speed, color, or whatever, so one population drifts gradually from another until the two groups are different enough to be considered separate species. Some may recall the story of Darwin’s finches: Similar birds on the Galapagos islands had different beak shapes, and Darwin concluded that the shapes matched the sort of plants that were available on each island. This example helped Darwin articulate his famous theory.

And that’s a fine story, but it answers the wrong question. We already know quite a lot about the type of genetic variation that might produce different beak shapes, because mankind has been selectively breeding animals for millenia. If you want to breed dogs, for example, then it’s no big trick to start with a group of dogs and selectively breed them for various traits. Given enough time and money, through selective breeding you can make them bigger or smaller, with longer or shorter fur, favoring certain fur colors or face shapes. But nothing in the animal breeder’s art will make them grow wings, or a second set of legs. Animal genes vary naturally, but only in certain ways.

Compounding the problem is that it’s hard to see how an animal would survive with a partial version of a feature that might eventually become useful. Some animals with severe deformities can survive and possibly even breed, as in this cat with two heads, or the various cat deformities listed here, but those hardly seem like promising branches of a new and distinct breed. The vast majority would die without special care. And many of those deformities are the result of problems in gestation, so they wouldn’t be inheritable. Microbiologist Michael Behe has made a name for himself refining this idea, which he calls Irreducible Complexity.

Beyond that, the gradualist approach to species evolution has problems with the fossil record. It’s a complicated subject, but basically we don’t find the smooth continuum from one species to the next that we would expect with gradual evolution. In fact, it seems that most experts on evolution do not agree with evolutionary gradualism, because of the fossil record. Instead, the current majority view seems to be Punctuated Equilibrium, which was first proposed by Niles Eldredge and Stephen Jay Gould in 1972. Many say that it’s just a fancy way of describing the fact that the fossil record is discontinuous, though others say that it’s much deeper than that. But the big point is simply that the formation of most new species did not occur gradually. However it happened, it happened quickly.

In the fast-evolution field, the best-known skeptic is William Dembski, who writes extensively about evolution from a mathematical perspective. Unfortunately, Dembski is also active in the political side of the evolution debate, which makes him an easy target for ridicule. It has also made him quite cynical about even the possibility of a rational debate on the subject, and that cynicism infects much of his writing.

So let’s leave Dembski aside and sketch from a blank slate some simple mathematical or information-theory objections to fast evolution.

The starting idea is that the development of a species requires the creation of information: At some point in time, there was no dog DNA. No cell anywhere on Earth knew how to make a dog. Then, at some later point in time, there were cells that contained dog DNA and knew how to make dogs. Somewhere along the way, the instructions for making dogs came into being.

How much information is involved in making a dog? There are various ways to answer that question, but they all lead to the same broad answer: A lot. Dog DNA must describe a particular dog, and must distinguish it from all the other things that DNA can make. And the more information that it takes to make a dog, the harder it is to see how dogs could have arisen by chance.

Let’s back up and sharpen the question a little. The non-political dispute over evolution comes down to a surprisingly subtle point: Did the DNA arise by chance, or was it created by intelligence? The mathematical evolution skeptics, like myself, have trouble with the chance option.

You don’t need anything more than high-school algebra to see the problem. Let’s take a simple example:

To be, or not to be–that is the question:
Whether ’tis nobler in the mind to suffer
The slings and arrows of outrageous fortune
Or to take arms against a sea of troubles
And by opposing end them.

There’s some information. It’s not much. It’s certainly less information than you would need to create a dog. Let’s run a few rough calculations on how long it would take for chance to generate that information randomly. And let’s push every assumption drastically in favor of generating it. Let’s strip out the punctuation and spaces; I count 153 characters. And let’s ignore capitalization. So all we need to randomly generate is a sequence of 153 characters, where each could be one of 26 options.

And let’s pour some resources into randomly generating that sequence. Let’s hire monkeys and put them at typewriters. These are fast monkeys, too. They push 153 random keys on their typewriters once every tenth of a second, year after year. And they don’t take banana breaks. They also have a foolproof detection mechanism: As soon as one of them types the target information, a computer will instantly snatch it up and let us know.

We don’t want our monkeys to be lonely, so let’s get a billion of them, at a billion typewriters. And let’s not crowd them on time, either. Let’s have them start at the Big Bang, and run continuously up to the present.

Is that generous enough? Is that enough time and speed for random monkeys to generate some information? Let’s find out.

First, let’s figure out how many random attempts our monkeys will make: The Big Bang happened about 13.7 billion years ago. That’s about five trillion days, 120 trillion hours, 7×10^15 minutes, 4.3×10^17 seconds, and 4.3×10^18 tenths of a second. Multiply by a billion monkeys, and you have 4.3×10^27 total attempts. Basically, a four with twenty-seven zeroes after it. And some very tired monkeys.

Surely, you might think, such a short sequence of letters would emerge thousands or millions of times amid so many attempts. And if you think that, then I strongly suggest Start-> Accessories->Calculator, then set to scientific mode. Because the next step is completely counterintuitive, and you probably won’t believe it unless you run the numbers for yourself.

To figure out how often those monkeys will bump into the sentence we want, we must calculate the size of the search space. Put simply: How many possible combinations are there for 153 letters, with 26 possible values for each letter?

I promise not to blab if you’ve forgotten your high-school algebra, but it isn’t hard: You start with the number of options per element, then apply an exponent equal to the number of elements. For one letter, there are 26 possibilities. For two letters, there are 26×26=676 possibilities, and so on.

Is your calculator open and ready? See what it gives you for 26 to the power of 153. (Use the x^y button in scientific mode.)

Got it yet? Go ahead. I’ll wait.

Anyone? Anyone? Yes, the hand in the back. That’s right. The answer is 3×10^216. That’s a three followed by two hundred and sixteen zeroes. I would type it out for you here, but frankly I don’t know what it would do to my blogging software. So use your imagination.

On each attempt, each monkey has a one in 3×10^216 chance of randomly typing out our little sequence of letters. Boy, that’s a low chance, isn’t it? But don’t worry: We’ve got a billion monkeys slamming out a billion new random sequences every tenth of a second for 13.7 billion years. That’s a lot of time, and a lot of monkeys. Won’t it all kinda work itself out?

Well, no. Because all those monkeys with all that time will “only” generate 4.3×10^27 random sequences, which is roughly a trillion times a trillion, four thousand times. It’s an awfully big number, even by US Congress standards. But it’s practically zero compared to the search space.

A number like 3×10^216 really defies comprehension. Let’s try to put it in perspective:

• There are about 9×10^49 atoms in the Earth.
• There are about 10^79 hydrogen atoms in the entire visible universe.
• There are about 10^88 photons in the visible universe.

Those are some big numbers, but they’re still nowhere near the number of possible combinations for a sequence of 153 letters. Nothing in the physical world comes close.

Was it unfair for me to pick a specific sentence? Fine. Let the monkeys try to produce any one of a billion sentences. Or a trillion! Take every 153-character English sentence that has ever been uttered by man, and then a googol more. It doesn’t matter, because in the end you’ll still have to divide by 3×10^216, and your odds of even one hit are lower than any lottery.

Does natural selection solve this? It’s hard to see how, because natural selection can only select something that already exists. The big problem isn’t with spotting the information once it arises; it’s with how the information could ever arise in the first place. And for random forces to produce every species on Earth, they must somehow generate information against these kinds of numbers over and over and over again.

I’ve presented this argument to reasonable evolution supporters, and the usual response is: “Well, but those are letters. DNA is different.” They never tell me how it’s different; they just assure me that it’s very different.

They’re right, of course. DNA is different: It’s far more complicated than four lines of Shakespeare. And no species has been producing ten billion mutations per second since the dawn of time. If the biochemists really have some special knowledge on how to generate gigabytes upon gigabytes of information without the intervention of any type of intelligence, then I wish that they would share it with the rest of us, because they’ve discovered the Holy Grail of Computer Science! It would be the biggest mathematical discovery since Calculus!

It’s while thinking of this that I occasionally lose my temper at those who smugly assert that anyone who doubts evolution is a knuckle-dragging, mouth-breathing, semi-literate moron eager to plunge our country into a Dark Night of Theocracy. Evolution, as popularly understood, assumes an unintelligent information-generation system that wildly surpasses anything that we can currently build—or even imagine! And we can do a lot of imagining inside of a computer these days. The problem isn’t that we don’t have enough computing power. The problem is that we don’t have any idea how to write the software. We don’t have any idea how anything other than a human being or similar intelligence could generate that volume of information.

Now I’ll shift gears and say that though the arrogance of most evolution supporters disgusts me, I do not subscribe to Intelligent Design, because I think that it goes a step too far. I am merely an evolution skeptic. The biological information around us did not arise by chance, but from that it does not follow that it was designed by intelligence. There are other options.

For all the chatter I’ve read about Intelligent Design, I’ve never seen a clear definition of ‘intelligent’. People are intelligent. God is intelligent, and many anti-IDers assume that ‘intelligence’ is just a synonum for ‘God’. But there are lots of other things that might or might not be considered intelligent. The boundary just isn’t very clear.

Maybe DNA has the capacity to change itself, like a self-programming computer. If the self-programming feature arose randomly, but all subsequent species development was the product of some algorithm interacting with the environment, does that count as random, or intelligent?

Of if, as one biochemist friend of mine believes, the molecules of biochemistry just happen to fit together such that a huge proportion of their combinations turn out to be useful, does that vindicate those who say that it’s random, or does that mean that the molecules themselves are intelligent?

So in the end there isn’t much substance to the non-political debate between random evolution and intelligent design. Both sides point to the black box that is the mechanism of species development, and they argue over which of the words ‘intelligent’ or ‘random’ better describes what’s going on in there. It’s more likely, I think, that the truth of the matter is simply beyond our grasp right now.