The Three-Body Problem

In 1687 Isaac Newton published the most successful equation in the history of science: $F = Gm_1 m_2/r^2$. With that one law he could predict the Moon's path around the Earth, the Earth's path around the Sun, and the return of Halley's Comet to the day.

The trick is that Newton's exact solution works perfectly for two bodies. The Earth and Moon, or the Sun and a lone planet — any pair of massive objects traces a perfect, repeating ellipse that you can write down as a formula.

Then someone asked: what about three? Add a third planet, a passing asteroid, or a second star — and the elegant formula evaporates. The three bodies tug at each other from all directions at once, the tugs change the distances, the new distances change the tugs, and the result spirals into chaos. No closed-form answer exists, and in 1890 Henri Poincaré proved it never will.

The simulation below integrates Newton's law of gravity for three bodies using the Runge-Kutta 4 method — the same family of algorithms used in real orbital mechanics software. Each body pulls the other two; nothing is simplified.

<p class="hint">{{hint}}</p>
<canvas id="c" width="480" height="360"></canvas>
<div class="btns">
  <button id="btn-launch" type="button">{{btn_launch}}</button>
  <button id="btn-perturb" type="button">{{btn_perturb}}</button>
  <button id="btn-reset" type="button" class="ghost">{{btn_reset}}</button>
</div>
<div class="status" id="status">{{status_initial}}</div>

* { box-sizing: border-box; }
body { font-family: system-ui, sans-serif; color: #222; margin: 0; background: #f4f6f8; }
.hint { font-size: .88rem; color: #444; margin: 0 0 .6rem; line-height: 1.45; }
canvas { display: block; width: 100%; max-width: 480px; border-radius: 10px;
         background: #0d0d1a; margin: 0 auto .5rem; }
.btns { display: flex; gap: .5rem; flex-wrap: wrap; justify-content: center; margin-bottom: .4rem; }
button { font: 600 14px system-ui; padding: .42rem .9rem; border: 1px solid #1d3557;
         background: #1d3557; color: #fff; border-radius: 8px; cursor: pointer; }
button.ghost { background: #fff; color: #1d3557; }
.status { font-size: .88rem; font-weight: 600; text-align: center; min-height: 1.3em;
          color: #1d3557; }

// Code not found

Click Launch to start with a symmetric figure-eight — one of the rare stable solutions discovered by Chenciner and Montgomery in 2000. Then click Perturb to nudge one body by a tiny amount. Within seconds the tidy choreography falls apart, and the orbits tangle in a completely different pattern. That extreme sensitivity to initial conditions is the signature of chaos: two trajectories that start almost identical diverge exponentially fast.

Why can't mathematicians just find the right formula?

The two-body shortcut. When two masses orbit each other, a change of variables converts the system into a single body moving around a fixed point. That body follows a conic section — an ellipse, parabola, or hyperbola — and every quantity can be expressed exactly in terms of the initial conditions. The solution is integrable: there are as many conserved quantities (energy, momentum, angular momentum) as there are degrees of freedom.

The three-body failure. Add a third mass and the count breaks down. Three bodies have 18 degrees of freedom (position + velocity, three dimensions each, three bodies). The system has only 10 conserved quantities. The missing constraints mean the motion is generically non-integrable — the trajectories can wander through a much larger region of phase space, and they do.

Poincaré's proof (1890). King Oscar II of Sweden offered a prize for solving the n-body problem. Poincaré entered with a series solution — and while reviewing his own work for publication, he found an error. Fixing it forced him to invent the concept of homoclinic tangles: regions where stable and unstable manifolds of orbits intersect in infinitely complex ways. His corrected paper became the founding document of chaos theory. The conclusion: no convergent power series in the initial conditions can give the positions for all time. The problem is provably non-integrable in general.

Chaos and Lyapunov exponents. In chaotic systems, the distance between two nearby trajectories grows as eᵏᵗ, where λ (the Lyapunov exponent) is positive. For typical three-body configurations, λ > 0: prediction horizon shrinks as you demand more precision, and doubling your measurement accuracy only buys you a fixed extra time before errors explode. This is not a failure of computing power — it is a mathematical property of the equations themselves.

Special stable solutions exist. The figure-eight orbit (Chenciner & Montgomery, 2000) and a handful of other choreographies are stable for small perturbations. But they form a set of measure zero among all possible initial conditions. For a randomly chosen starting configuration, chaos wins.

Related reading: the Lorenz attractor explores chaos in fluid dynamics using the same sensitivity ideas; n-body simulation covers the numerical algorithms used to integrate many-body gravity.

The absence of a formula is not just a theoretical inconvenience — it shapes entire branches of science and engineering:

• Spacecraft trajectory design: missions like the James Webb Space Telescope use Lagrange points, the five equilibrium positions in a Sun-Earth three-body system. Getting there requires months of numerical integration to find a transfer trajectory that uses gravity assists efficiently.
• Solar system stability: is the solar system stable over billions of years? Numerical simulations (Laskar, 1989) show that Mercury could be ejected within five billion years — a three-body chaos effect. No formula can answer this; only long simulation can.
• Binary star systems: when a third star passes through a binary pair, any of the three bodies can be ejected. The outcome is chaotic and must be sampled statistically over thousands of simulated encounters.
• Galaxy dynamics: galaxies are n-body systems with $10^{11}$ stars. The three-body problem is the building block — N-body codes like GADGET split the force calculation into short-range (direct three-body-like) and long-range (tree/mesh) components.
• Poincaré's legacy: the chaos theory he discovered to describe why three bodies have no formula became a tool for understanding weather, fluid turbulence, population dynamics, and the double pendulum you can explore at double pendulum.

Newton gave us a law that predicts the universe. But his own law breaks — not because of rounding errors or insufficient computers, but because of mathematics itself.

Three gravitating bodies form the simplest system that is provably non-integrable. Add a third mass to a solar system and the orbits can interleave, exchange energy, and eject a body entirely — and no formula will ever tell you which outcome to expect from a given starting point without running the simulation.

That is Poincaré's legacy: the search for a formula for three planets accidentally gave us chaos theory, one of the most powerful ideas in all of science. The next time you see a space mission trajectory or a weather forecast, remember — the reason it is a simulation and not a formula goes all the way back to three bodies pulling on each other in the dark.