Beyond the Arithmetic
Up to this point, this book has worked almost entirely with numbers. We counted drums on an odometer. We asked how many tries every bacterium that has ever lived could have made. We timed a single small stretch of DNA in the human line, and watched a word game stall when its steps had to stay on an island of sense. Each time, the question was the same: given everything the scientists say, does the arithmetic work? Each time, the answer came back that it doesn't, and it didn't come back as a close call.
The rest of this book takes a different approach. The chapters ahead look at some of the components of life itself: a few of the machines a cell can't do without, and a few things that go far beyond any single cell, instinct and the brain. I can't attach precise numbers to these. Nobody can. They're too large, too intricate, and too tangled up with everything around them for any honest odometer. What follows is evidence, not arithmetic: what biologists have actually found, explained in plain language as we go.
But the arithmetic doesn't disappear. It sets the standard everything else has to be measured against.
Remember what it took to find a single working protein of three hundred amino acids. Even on the most generous assumptions, everything the universe could ever offer fell short. Remember how badly point mutation fares inside a gene that already works: it's a demolition tool far more often than a building tool. Remember how quickly a step-by-step search breaks down when its progress can't be locked in. A protein needing twenty precise positions was already at the edge of what bacteria could find, and one needing forty-two was beyond it.
Now consider what's coming. Some of the machines in the chapters ahead are built from many proteins at once, several of them far longer than three hundred amino acids, and all of them have to fit together and work as a unit. And these aren't late additions, refinements that appeared once life was well established and had billions of years to tinker. They're fundamental. As far as anyone can tell, they had to be there very early on, because without them nothing else works.
By the standards this book has already established, such things are impossibly large. If a single modest protein is beyond the reach of chance, a machine made of dozens of interlocking ones is not just a harder version of the same problem. It's a different problem in kind, the same way the complete works of Shakespeare weren't just a longer sonnet.
One objection deserves an answer in advance. Earlier, this book insisted that the target is usually wide: many sonnets count, many sensible strings count, and many versions of a protein might do the same job. Couldn't the same be true of these machines? Perhaps, for any one part considered on its own. But a machine isn't a pile of parts. Each piece has to fit the pieces around it, and a looser fit for one part does nothing for the whole unless every part it touches loosens with it. Widening the target for each piece separately doesn't widen it for the assembled machine. If anything, the need for everything to fit together narrows it again.
So here is the challenge, stated as simply as I can. Any theory that holds that chance supplies the raw material for evolution has to explain how such large, fundamental machines were found, and found so early. "Step by step" isn't an answer by itself. It depends on long stretches of time, and on showing that each machine could have been built in smaller pieces, each one useful along the way, each one worth keeping. That second part is the hard part, and for the machines ahead there's no evidence that it happened. There's only the assumption that it must have.
The chapters that follow look at what that assumption is being asked to carry.
Last revised 2026-11-18