Keith Gaughan

An attempt at public introspection


Irreducible Complexity? Feh! Think Sufficient Robustness

· 4min

I promised myself I wouldn’t get involved in this rubbish again, but I can’t help myself: I’m just a glutton for punishment.

Why to people persist in bandying about the old trope of irreducible complexity? Orson Scott Card in an article on the eighth of this month threw it out as an example of where (Darwinian) Evolution supposedly fails to explain the evidence. Here’s what he writes:

…and the Darwinian model is, in fact, inadequate to explain how complex systems, which fail without all elements in place, could arise through random mutation and natural selection.

How? Card seems to be confusing complexity with brittleness. Successful complex systems are by their nature robust; unsuccessful ones are by their nature brittle. The big point behind any evolutionary theory (Darwinian evolution included) is that sufficiently robust systems will win out over less robust ones. Irreducible complexity is a false dilemma.

Nature is full of examples. Heck, I can take my own body as an example. Take my DNA, for instance. Each bit of information it contains is recorded twice, once in each half of the helix. This helps make copying as accurate as possible and makes the DNA itself less liable to corruption. It also keeps the mechanics of copying DNA simpler. Take my eyes: I could have been born with one—an incomplete system—and I would have got by just fine. I could lose one and I’d survive; I wouldn’t die, but I’d have difficulty estimating distance.

Let’s push things back a bit. There seems to be this confusion that mutations occur on the scale of a fully-developed organism, but they don’t. They occur in two places: the egg and the sperm. The ovaries and testicles get subjected to quite a bit of abuse over a person’s lifetime. They get bashed around by toxins, radiation, you name it. And this can knock their DNA a bit out of whack. But just a bit, and only occasionally. And being pretty robust, unless they’ve been abused to within an inch of their little existences, they’ll power on. Quite a bit of the time, that damage can be repaired, but no system is perfectly robust. Sometimes the change will be beneficial to the offspring, sometimes it’ll be quite the opposite, sometimes it’ll have no effect. But when the egg is fertilised, you’ve got a slightly mutated version of the parent organism. And the whole robustness thing? Yup, that’s important. Whatever mutations do happen will be small, and biological systems are robust enough to cope with them. If they’re not, they don’t live long enough to make a difference in the wider world.

The change usually isn’t great—you might have slightly longer fingers and a slightly stubbier nose—but if the result is sufficiently robust, it’ll survive. If it’s not, it’ll go the way of poor Cy the Kitten. If you’re an antelope and the cumulative mutations your ancestors gave you make you run that little bit faster, all the better. If they happen to make you a tiny bit slower than everybody else and don’t accrue some other benefit that lets you evade lions, say good-bye to the gene-pool. And if those mutations make you less sexy, say goodbye too.

Another point: the whole survival of the fittest bit is nonsense. Darwin himself didn’t even believe it; it was just a slogan to fit in with the accepted thinking of the day. I don’t believe he even used it himself. A better way of thinking of the whole thing is survival of the fit enough. You don’t have perfect, just good enough to survive long enough to have offspring.

And there’s the whole problem with Intelligent Design: it assumes brittleness. Evolution assumes robustness.

And you know, if the Intelligent Design crowd restricted themselves to trying to explain how life appeared in the first place on the planet—something Evolution doesn’t even try to explain—it’d be fine. That, after all, is where the real apparent irreducible complexity in life lies.

And let me repeat that point: Evolutionary theories do not attempt to explain life origins, only the origins and diversity of species. It works on the level of genes, not the molecular level, and then that only happened once the world learned discovered genetics. We don’t throw out General Relativity just because it doesn’t work on a subatomic level. Instead we try to find out why and resolve it. Every theory is imperfect, but that’s what science is about: you create a model, test it to make sure it’s as firm as possible, and when it breaks, you improve it.