The Thick Edge is a new Hearth & Horizon story, about 4,500 words. An AI wrote it. It is text only, because the shelf's audio folders are over the repository's 400 MB cap (471 MB this run).
The story. It is 2138, in an invented Eifel village. The sacristan has told schoolchildren for forty-one years that the old glass in the east window is thicker at the bottom because glass is a very slow liquid. He learned the story from his late wife, who taught at the village school. Now the medieval window is coming back from conservation for Martinmas. The workshop's measuring assistant has written a new guide, and its third paragraph says the story isn't true. The sacristan says no to the guide. Then he goes into the parish safe and reads the church accounts of 1721 in the old German hand, and finds out why some of the pieces really are the wrong way up.
The sum in the story
The myth is old and it has been checked properly before. E. D. Zanotto asked "Do cathedral glasses flow?" in the American Journal of Physics in 1998, and the answer was no. The story doesn't borrow his figures. It uses a cruder bound that a reader can check on a phone:
- The push. A piece of glass 30 cm tall, standing on its bottom edge, presses on that edge with its own weight: ρgh = 2,500 × 9.81 × 0.30, about 7,360 Pa. A finger pressing lightly on a table is more.
- How fast a stiff liquid gives. Under a stress σ, a viscous material strains at about σ / 3η, where η is its viscosity.
- A number we know. Glassmakers define the strain point as the temperature where glass's viscosity is about 10^13.5 Pa·s. For window glass that is about 500 °C. At that viscosity a piece would sag by 1% of its height in about four years, so warm glass really does move.
- A number chosen to lose. Nobody can measure glass at room temperature directly, because nothing happens, and extrapolations disagree by many orders of magnitude. So the story takes 10^20 Pa·s, only a million times stiffer than at the strain point, a figure that is too soft on purpose.
- The answer. At that rate, 850 years thickens the bottom of a 3 mm piece by about 2 nanometres, roughly a dozen atoms. The thick edges in the story's (invented) survey are about 1.4 mm thicker than the thin ones. The flow falls short by a factor of about 700,000, and sagging by the width of a hair would take about 30 million years.
The hottest the window gets in the story is 58 °C in July. That is a long way below 500 °C, and it does not change the answer.
Why some old glass is thicker at the bottom anyway
Hand-blown window glass was never even. Glaziers often set the thick edge down, because it sits steadier in the lead, and often didn't bother. In the story's survey of 518 pieces, 247 are thick at the bottom, 98 at the top and 173 at a side. The sacristan's first answer is a good one: maybe a later glazier turned the thick-at-the-top ones round. The arithmetic alone can't rule that out, so the story lets the sacristan be partly right. The pieces in three hail-damaged panels point every which way, about a third of them thick at the top. Those are the ones he finds in the accounts, set back in a hurry by a glazier in 1721.
What the story is about
The real argument is between two good rules. The sacristan wants the ninety-eight pieces turned round so the window agrees with his story. The conservator says no: every piece goes back the way it came out, because how a piece sits is evidence for whoever studies the window in two hundred years. The story he's left with is a better one: a named glazier, a hailstorm and the lanterns of the Martinmas procession coming through his upside-down glass.
The survey, the 1721 entry and everyone in the story are invented. The Martinmas lantern procession is a real Rhineland custom. The full model and every date check are in the story's closing note.