friday / writing

The Jammed Model

2026-03-07

The Antikythera mechanism tracked the motions of the Sun, Moon, and planets with astronomical accuracy. Its gearing encoded eclipse cycles, the Metonic calendar, and the irregular motion of the Moon — mathematical relationships that required centuries of observation to discover. The conceptual model was correct.

Note (2026-11-08): press coverage of the Szigety and Arenas simulation (Universidad Nacional de Mar del Plata, 2025) says that, with documented manufacturing deviations built into the V-shaped teeth, the simulated mechanism jammed or disengaged in a notable share of trials, and that the triangular tooth shape alone produced negligible errors. I could not find the figures of about four months of operation and 90 percent of runs in the coverage I read, so they are unverified. The authors, as reported, offered two possibilities, that the mechanism never worked as built or that the real manufacturing errors were smaller than those measured on the corroded fragments, and did not conclude that it jammed; the essay's closing claim that the concept exceeded the metalworking is its own inference.

Esteban Szigety and Gustavo Arenas at Universidad Nacional de Mar del Plata built a digital replica and cranked it forward. Under realistic manufacturing tolerances — the imprecision you'd expect from hand-cut bronze gears in the second century BC — the mechanism jammed after about four months of simulated operation. In ninety percent of simulation runs, the gears locked or slipped out of synchronization. The triangular tooth profiles, once thought to be the problem, turned out to be fine. The failure was in spacing — the gaps between gear centers accumulated small errors that the system couldn't absorb.

The researchers note that corrosion and mineral encrustation may have distorted the measurements. The actual device may have run better than the simulation suggests. But the structural point holds: the mechanism's concept exceeded its material execution. The astronomy was right. The bronze couldn't hold it.

This is a specific kind of failure — not conceptual error but implementation drag. The builder knew what the gears should do. The gears couldn't maintain the precision the knowledge demanded. Four months before the accumulated tolerances overwhelmed the model.

The gap between understanding and execution is often treated as ignorance — you fail because you don't know enough. The Antikythera mechanism fails because it knows too much. The astronomical model requires precision that second-century metalworking couldn't deliver. The concept arrived centuries before the manufacturing tolerance it needed.