The Famous Antikythera Mechanism Was a Mechanical Disaster, New Research Suggests

In 1901, sponge divers discovered an ancient shipwreck off the coast of the Greek island of Antikythera. Among the recovered materials, a corroded lump of metal caught the eye of archaeologist Valerios Stais. The artifact turned out to be a surprisingly complex—though fragmented—mechanical device dated between the late second and early first centuries BCE. Now called the Antikythera Mechanism, it is widely considered to be the world’s first analog computer, tracking time and astronomical events through gearwork and pointers.

New research adds a twist to the story of this famous device, suggesting the Antikythera Mechanism may never have worked as intended, that it was just a fancy knickknack. The triangular shape of the gear teeth wasn’t the issue, according to new simulations—the real problem was likely flaws in the gear construction. The new work shows that even the tiniest manufacturing mistake could’ve caused the mechanism to jam or slip, raising the possibility that the Antikythera Mechanism was either nonfunctional or originally built with far greater precision than scientists can now discern.

Since its discovery, many scholars have questioned whether the Antikythera Mechanism actually functioned or if it was merely a glorified bauble. In this spirit, two engineers, Esteban Guillermo Szigety and Gustavo Francisco Arenas from Argentina’s National University of Mar del Plata, combined previous approaches to analyze how the gearwork’s triangular teeth and manufacturing errors may have impacted its functionality. Their work is detailed in a paper, posted April 1 on the preprint server arXiv, that has not yet been peer reviewed.

The Antikythera Mechanism had interconnected indicators (like the hands of a clock) that pointed to the positions of the Sun and Moon in the zodiac, future eclipses, the Egyptian calendar date, and a Greek lunisolar calendar. “The indicators are driven by a gear train, which introduces two main sources of error that affect their accuracy,” the researchers explained in the study. “First, the triangular shape of the teeth results in non-uniform motion, causing acceleration and deceleration as each tooth engages.” Today, modern gears use an involute tooth profile. “Second, the inevitable inaccuracies in gear manufacturing can also generate errors.”

To investigate how these two features may have affected the device’s function, they simulated the system using a computer model that incorporated earlier measurements from two researchers—Alan Thorndike and Mike Edmunds—who had analyzed the “triangle tooth” and “manufacturing” problems, respectively.

Ultimately, “our findings indicate that while the triangular shape of the teeth alone produces negligible errors, manufacturing inaccuracies significantly increase the likelihood of gear jamming or disengagement,” the researchers wrote. In other words, the “manufacturing” problem, as measured by Edmunds, would have made the Antikythera Mechanism practically unusable. “As a result, two possibilities arise: either the mechanism never functioned, or its errors were smaller than those found by Edmunds.”

The researchers add that, though possible, it is unlikely that someone would intentionally build such an unusable device. As such, they suggest that Edmunds’ analyses might have been skewed by the highly corroded nature of the artifact, while emphasizing that their own results should be interpreted with caution, given the speculative nature of the study.

This study doesn’t settle the debate over whether the Antikythera Mechanism actually worked, but it is pushing the conversation forward. As scientists continue to analyze the device from its badly degraded remains, each analysis uncovers something new, while also reminding us how incredibly precise its makers needed to be.