Why You Can Cut a Ball Into Two Identical Balls

In 1924, Stefan Banach and Alfred Tarski published a paper proving that a solid ball can be cut into a finite number of pieces that reassemble, using only rotations and translations, into two solid balls each the same size as the original. The pieces are not bent, not stretched, not duplicated. Just rigidly moved.

The minimum count is five. Raphael Robinson nailed this down in 1947: four pieces are not enough, and five suffice. So in principle, you could partition a billiard ball into five subsets, rotate each, and end up with two billiard balls.

In principle. Try it physically and the trick collapses. The five pieces are not chunks of ball — they are infinite scatterings of individual points, dense in a way that no knife or laser can separate. They are also non-measurable: ask what volume any single piece has, and the question has no answer. Lebesgue measure, the standard notion of length, area, and volume that underlies calculus, simply does not apply to them. Without a defined volume per piece, the usual arithmetic — pieces add up to the volume of the whole — has nothing to add up.

The construction relies on the axiom of choice, the rule that lets you pick one element from each of infinitely many non-empty sets even when no procedure tells you which. In 1964, Paul Cohen proved choice is independent of the rest of the standard axioms, so a mathematician can take it or leave it. Take it, and you get the paradox. Drop it, and the spheres stay one. The strange object Banach and Tarski conjured was never a flaw in geometry. It was the price of an axiom most working mathematicians keep, because dropping it costs more than it saves.