Do old cathedral windows flow like a liquid (glass flow myth)

Settled in 1998 by Edgar D. Zanotto (Federal University of São Carlos, Brazil).

The claim that glass is 'really' a very slowly flowing liquid, popularly illustrated by medieval cathedral windows being thicker at the bottom, has circulated since at least the early 20th century and is often repeated as a fun fact even in some chemistry textbooks. In 1998 Edgar Zanotto calculated the actual viscosity of both modern and historically representative ancient window glass compositions at room temperature and estimated the relaxation time over which any observable flow could occur, finding it to be on the order of 10^23 to 10^32 years — vastly longer than the age of the universe. This showed decisively that any glass flow at room temperature is utterly negligible on human timescales, and that the uneven thickness of old windows instead comes from the imperfect crown- or cylinder-glass manufacturing methods used before modern float glass, with installers commonly placing the heavier, thicker edge at the bottom.

A charming, widely repeated 'secret liquid' story about ancient windows collapses the moment someone actually calculates glass's viscosity, which turns out to dwarf the age of the universe.

Explanations that circulated

• Medieval window glass is thicker at the bottom because the solid glass has visibly flowed downward under gravity over centuries — debunked; the uneven thickness is instead an artifact of the imperfect glassmaking (crown- or cylinder-glass) methods of the era, and old glaziers often installed the thicker edge at the bottom for stability, not because the glass moved after installation.

References

• Do cathedral glasses flow?, E. D. Zanotto, American Journal of Physics, 1998