What is Irreducible Complexity?

I received an email from a fellow podcaster, Emery Wang, who wrote to ask me about the arguments from scientists who deny evolutionary theory to be scientific. To be blunt, there really aren’t that many. The best place to find them, to the best of my knowledge, is the Discovery Institute, which is a creationist organization that tries its very best to portray itself as scientific. You can find them online at www.discovery.org. They do actually have a few scientists in their ranks, but only a couple with the necessary biological credentials to speak authoritatively about evolution from a scientific standpoint. Their shining star, so to speak, is a man named Michael Behe, although most of the other well-known names in the Intelligent Design movement, such as William Dembski and Jonathan Wells, are also affiliated with this group. Behe is a legitimate scientist, a biochemist actually, and a professor at Lehigh University in Pennsylvania. If you know Behe at all, you know him primarily as the author of the book, “Darwin’s Black Box: The Biochemical Challenge to Evolution.” It’s the book that has the old man and the chimpanzee sitting next to each other, facing in different directions. It’s in this book that Behe first put forth the idea of “irreducible complexity.”

Irreducible complexity was an incredible success for the Intelligent Design movement, because at its core it’s very intuitive- unfortunately, however, it’s also unscientific. But since the scientific theory of evolution is counter-intuitive, it was widely accepted by the public, while scientists who knew better were grinding their teeth in frustration.

Irreducible complexity works like this: Suppose that you have a mechanism of some sort, perhaps a mousetrap. This mousetrap is composed of several different parts, each of which is essential to the operation of the mousetrap. You’ve got the flat wooden base, the spring, the horizontal bar, the catch bar, the catch, and the staples that keep all the metal parts attached to the wood base. Now, if you have all the parts together and assembled properly, the mousetrap works like it’s supposed to- pulling back on the horizontal bar causes the spring to wind back, and the catch bar holds the horizontal bar in place as long as it’s jammed in place by the catch. Once the catch is disturbed, the catch bar is free to swing out of the way, and the spring winds shut slamming the horizontal bar down hard on whatever disturbed the catch. Makes sense. But let’s say that you remove one part of the mousetrap- the catch. Well, in that case, you can never set the trap because you can’t keep the catch bar still. Or let’s say that you remove the spring. Well, in that case, the trap will never close because there’s no force to move the horizontal bar. Or if you remove the horizontal bar itself, there’s nothing for the spring to move. You get the idea, I’m sure- if you remove one part of the mechanism, the whole thing can’t work. Thus, the design of the mousetrap is described by Behe to be irreducibly complex- in other words, the complexity of the design requires that it can’t be reduced any farther without losing functionality.

Now, you’re thinking, “So what’s the problem with irreducible complexity? Obviously a mousetrap won’t work if you remove the spring, that’s just common sense.” And so it is. And I don’t think there’d be any problem if Behe had stuck to talking about mousetraps. But he doesn’t- he’s a biochemist, and so he applies this concept of irreducible complexity to something much smaller than a mousetrap, something so small it’s invisible to the naked eye- a bacterial flagellum. A flagellum is a long, whiplike structure that is used by cells to move around. Think of the tail on a sperm cell, and you’re basically there. Bacteria use flagella too, and the structure is of a long, hollow cord attached to the wall of the bacterium where it hooks into a molecular rotor that spins in response to an ion gradient, as much as 1000 rpm. When the rotor spins, the flagellum spins, and the bacterium moves forward. It’s a little bit like an outboard motor on a boat, and like a motor, it’s composed of a lot of different parts, in this case proteins, each of which is essential for the proper functioning on the flagellum. Well, you can probably see where I’m going with this: if the mousetrap is irreducibly complex since removing one part means that the whole thing doesn’t work, then a flagellum is irreducibly complex for the same reason, right?

Wrong. And hopefully once I explain why, you’ll understand that it’s necessary to assume design in order to come up with a concept like irreducible complexity in the first place. Let’s go back to the mousetrap. Let’s say that I remove the catch- mousetrap doesn’t work, right? Well, not exactly. It may not work the way that the manufacturer intended for it to work, that’s true, but is it absolutely good for nothing? I would say no- I can set the trap by pulling the horizontal bar into place, setting the catch bar over it, and then carefully laying the trap upside down so that its own weight holds the catch bar in place. I can still bait the trap, and the jostling of the upside-down trap by an eager mouse can still move the catch bar out of position and cause the trap to release. It’s not the best way for the trap to function, of course. But even though it’s not as effective, it still does function to some extent. And even if you removed the spring, you could still use it as a paperweight- it’s still good for something.

Likewise, the incomplete flagellum is also good for something. Recent research on bacterial flagella have shown that very similar proteins functioning in a very similar way, but without the flagellar whip structure, have another kind of function in bacteria- they form the basis of a secretory apparatus- a mechanism that allows bacteria to inject toxins into other cells. So half a flagellum is still useful to the bacteria, even if it’s not functioning in the same way that the full flagellum does. Thus, the flagellum is not irreducibly complex. You see, evolutionary theory doesn’t have a particular goal in mind- that’s why Behe’s analogy of the mousetrap doesn’t make any sense. Someone who sets out to build a mousetrap has an idea in mind of what he wants that trap to be capable of, but this is not the case for evolution. Structures and systems are only useful to an organism if they confer some kind of selective advantage- it doesn’t matter how that advantage operates.

Think of a fighter jet- that’s a pretty complex system, right? Every component in that fighter jet is absolutely essential for its operation, otherwise it wouldn’t be built into it. If you were looking at the jet from Behe’s point of view, you would say, wow, this jet is incredibly complex. This must have been engineered from a blank diagram specifically to work this way. But we know better. Airplanes themselves have been constantly evolving, since even before the Wright brothers flew their machine in North Carolina. Before that, there were all sorts of variations on gliders. Each component was added gradually, but each airplane that was built started with the blueprint of the one that went before it, and then added new things. If those additions made a better airplane, then everyone copied it. If those additions made it worse, then it was scrapped. This is how evolution works- variations are made on existing organisms, and if they confer an advantage, they are selected for. Just as we know that an airplane motor evolved from a more simple model, we can also show that a bacterial motor, a flagellum, evolved from a more simple structure.

And in fact, after looking at it in both simple mechanical and biomechanical examples, what does the concept of irreducible complexity really give us? Not really anything useful at all. Because the critical component here is complexity- this is something which is dependent on the proposed function. A mousetrap is a reasonably complex way to kill a mouse, but it’s a simple enough paperweight at the same time. And a bacterial flagellum is a reasonably complex way to move a cell around in a fluid medium, but it’s also a reasonably good way to inject proteins into other cells, if you take a few parts away. And complexity is also dependent on the point in time which you examine a system. If you look at a stone arch, it seems irreducibly complex- if you remove a single stone from the archway the whole thing comes tumbling down. But we know that a common architectural technique is called scaffolding- that is, the archway is built under support from a wooden scaffold that allows the stones to be put in place without falling apart- and it’s only after all the stones are secure that the scaffold isn’t needed anymore. Biological systems can evolve using scaffolds also- with less selective pressure, new proteins and enzymes can evolve unique functions which may become essential if other scaffolding enzymes are lost to the dustbin of evolutionary change.

So, in the end, we’ve seen that irreducible complexity is neither- biological structures are, in fact, reducible to states which give rise to other functions, and these functions are only as complex as their context requires of them. It’s an attractive concept to people that aren’t familiar with evolutionary theory, but it’s just not born out by science.