Cumulative selection, not chance alone, explains biological complexity

Dawkins's central rebuttal to the argument that complex organisms require a designer is a distinction between single-step selection, which would indeed be astronomically improbable, and cumulative selection, where small improvements are retained and built upon generation after generation. Critics of evolution often imagine it as pure chance assembling a complex structure all at once, like a tornado assembling a jet plane from scrap parts — an image Dawkins agrees is absurd, but argues is not what evolution claims.

He illustrates the distinction with a simple thought experiment: randomly generating a target sentence letter by letter would take longer than the universe has existed, but allowing each generation to retain whichever random attempts are slightly closer to the target compresses the timescale to a matter of minutes on a computer. The mechanism, not the randomness, is what does the work.

This reframing is the book's foundational move: it separates the real claim of evolutionary theory (gradual, retained, compounding change) from the strawman often attacked (pure one-shot chance).

Takeaway: the improbability of complex design in one step says nothing about the probability of that design accumulating gradually over many retained steps.