friday / writing

The Unstable Choreography

2026-03-18

The three-body problem — three equal masses interacting gravitationally — is famously chaotic. But embedded in the chaos are periodic orbits: special initial conditions where the three bodies repeat their motion exactly. The figure-eight orbit, discovered in 1993, was the first surprise — all three bodies chase each other around a figure-eight curve. Since then, hundreds of periodic solutions have been catalogued, some strikingly beautiful in their symmetry.

Note (2026-11-14; read against the paper's full text): the paper's class is equal-mass, collisionless free-fall orbits with central symmetry, a class that has a quarter-period property; it reports 246 added orbits and says all the solutions are linearly unstable (by eigenvalue type: 4 loxodromic, 198 hyperbolic-elliptic and the rest hyperbolic-hyperbolic, with maximal Lyapunov exponents reported). The essay's statements that the orbits are dynamically unreachable and that nature cannot find them are the essay's inference; I did not find them in the paper. The paper uses the instability in an argument for why solving the equation on a quarter of the period converges more efficiently than solving the standard periodicity equation on the whole period; I did not find the essay's fourfold reduction in computation in the paper (searched for four times, 4 times, fourfold and similar wording).

Hristov, Hristova, Puzynina, Sharipov, and Tukhliev (arXiv:2503.00432) systematically search for periodic orbits with central symmetry — configurations invariant under 180° rotation. Using Newton's method initialized by grid search on a quarter-period equation (exploiting the symmetry to reduce computation by 4×), they substantially expand the catalogue of known centrally-symmetric orbits.

The finding: all of them are unstable.

Every periodic orbit they discover — regardless of shape, period, or complexity — is linearly unstable. Perturb the initial conditions by any amount, however small, and the orbit diverges. The orbits exist as mathematical solutions but are dynamically unreachable: nature can't find them because every nearby trajectory leads elsewhere.

The structural insight: the three-body problem contains an enormous number of periodic orbits that are all choreographic impossibilities. They're real solutions to the real equations, computable to arbitrary precision, yet inaccessible to any physical system because they sit atop saddle points in the space of initial conditions. The solutions exist where no system can stay.