Chapter 6: The Monkeys That Didn't
Nine Versus One
Now, back to the question that matters for an understanding of evolution: does working in small chunks instead of single letters make the underlying problem any easier? This is worth answering carefully, because the intuition that it should is exactly the intuition evolutionary theory leans on when it talks about small, incremental steps.
Go back to the odometer. Building a three-hundred-letter target one single letter at a time needs three hundred drums, each holding the twenty-six letters of the alphabet. Building the same target nine letters at a time instead needs roughly thirty-three or thirty-four drums — far fewer drums, which sounds like real progress — except each of those drums now has to hold not twenty-six options, but all 5.4 trillion possible nine-letter combinations. Work through the multiplication, and the two numbers come out the same. Not approximately the same — the same, because they're two different ways of writing an identical calculation: twenty-six multiplied by itself three hundred times, regrouped into chunks of nine instead of chunks of one, changes nothing about the total. The same goes for proteins: three hundred ninety-one drums, whether the amino acids arrive one at a time or nine at a time. (The arithmetic: Nine at a Time.) Smaller chunks don't shrink the mountain. They just change how many steps you take to describe climbing it.
So the answer to the question at the start of this chapter is no. Groups of nine don't improve the odds at all.
There's a second problem, specific to the origin of life, that makes chunking even less helpful than the arithmetic alone suggests. Chunking only helps a search if there's some mechanism keeping the good chunks and discarding the bad ones as you go — exactly the job natural selection is supposed to do, once life already exists to do the selecting. But before life exists, there's nothing to select for. A nine-letter fragment of protein floating in a prebiotic ocean has no functioning cell to keep it, use it, or prefer it over any other equally useless nine-letter fragment sitting right next to it. Chunking helps if something is choosing between chunks. Before life exists, nothing is.
And there's a third wrinkle worth adding, because it undercuts the whole nine-letter choice from a different angle entirely. There was nothing principled about Anderson landing on nine letters specifically. Each additional letter in a chunk multiplies the number of possible chunks by twenty-six. At ten letters, the job that took Anderson about six and a half weeks would have taken more than three years; at eleven, more than eighty. At twenty letters, the pool of possible chunks is nearly four quadrillion times larger than the nine-letter pool Anderson actually used, and the same job would take more than thirty thousand times the age of the universe. (The arithmetic: Bigger Chunks.) Not a modest slowdown, but a wall. There's nothing magic about the number nine. It was simply small enough to finish.
That points to a trap, and it's worth naming now. Small chunks are easy to find, but they're too short to do anything: as we saw in Chapter 3, a working protein machine generally needs at least a hundred or so amino acids. And a chunk that does nothing gives selection nothing to choose. Chunks long enough to do something, on the other hand, come in so many possible versions that there's no finding the right ones. Make the pieces small, and they're useless; make them big enough to be useful, and they can't be found.
These are all things worth keeping in mind when we turn, in a later chapter, to evolution proper — once life has already begun, and there's finally something in the picture that can do the choosing that chunking depends on.
Last revised 2026-11-17