The Slow Dial
In 2702 the sundial on a Görlitz church has read about twenty-four minutes slow, even with its own table applied, for as long as anyone remembers, and the old stonemason asked to replace it finds his new dial slow by the same amount. The Earth's day has grown about twelve thousandths of a second longer since the clocks stopped adding leap seconds in the 2030s, the shortfall has added up, and an apprentice works out why a dial cut to agree with the clock would be wrong again within a century.
Listen (34 min, synthetic voice: Kokoro)
Or subscribe to the Hearth & Horizon podcast feed in any podcast app that accepts an RSS link.
Lutz Hennig had made sundials in Görlitz for fifty-one years, every one of them right, and for fifty-one years he had despised the one on the south wall of Saint Hyacinth's.
Everyone in the town knew it as the slow dial. It was a slab of grey sandstone a metre and a bit high, with an iron style sticking out of it at an angle and the hours cut in Roman numerals in a fan below, and a table cut in a second stone under it. On any sunny day of the year it was behind the clock on the town hall: by seven or eight minutes in early November, by well over half an hour in February, and, once you had worked the table, by about twenty-five minutes all year round. Children were sent to look at it as a joke. Tourists from the Polish side of the river photographed it next to their watches. At the bottom, under the motto ULTIMA LATET, the maker had cut his name and the year, M. SOBCZAK 2051, very neatly, and Lutz had walked past that name on his way to Mass for half a century and thought, every time, a man who could cut letters like that and could not set a style.
So when the parish decided at last to restore the south wall, and Father Chidi came to the workshop on a wet afternoon in April to ask whether Lutz would make them a new dial, Lutz said yes before the priest had finished the sentence.
"It must be right," Father Chidi said. "That is the only thing the committee asked. Frau Kästner has raised the money on the promise that it will be right. People want to be able to look at it and set their watch."
"It will be right," said Lutz. "Sobczak did not know what he was doing. I do."
"And you will take the girl with you on it? Frau Kästner says she is very good."
The girl was standing at the bench behind him with a mallet in one hand, listening and pretending not to. Agnieszka Wróbel was seventeen, came over the footbridge from Zgorzelec every morning at seven, and had been his apprentice for two years. She was very good. She was also the only person who had worked for him in fifty-one years who argued, and he had kept her on partly for that and partly because his hands, since the winter, had begun to shake when he held a fine chisel for more than a minute at a time. He did not talk about the hands.
"The girl will do the arithmetic," he said. "I will do the dial."
"I will do the arithmetic by hand," said Aga, "so you can check it, and then I will ask Bede to check it, so we can both be sure."
Lutz made the noise he made about Bede.
Görlitz sits on the fifteenth meridian east. That is the line Central European Time is counted from: when the clocks were set up, a clock in Görlitz at noon was meant to be a clock at the moment the Sun stood due south, give or take the wobble the almanacs call the equation of time, which is never more than about a quarter of an hour either way and comes back to nothing four times a year. The town had marked the meridian for longer than anyone remembered, and was proud of it, in the way small towns are proud of the one thing about them that is in the encyclopaedias. It is a strange thing to be famous for being exactly on time, and then to have, on the wall of a church in the middle of town, a dial that is not.
Lutz worked from his grandfather's book. It was a real paper book, printed in the 2040s, with a cracked spine and his grandfather's pencil in the margins, and it said, in a sentence Lutz knew by heart: A dial correctly made shows local apparent solar time. To obtain clock time, apply the equation of time from the table, and the longitude correction; on the fifteenth meridian the longitude correction is zero.
"So we do not have even the longitude," he said. "Görlitz is the easy case. A south wall, the fifteenth meridian, a table. A child could make this dial."
"A child is making it," said Aga, who was working out the hour lines on the back of a delivery note.
The geometry of a dial on a wall facing due south is old and simple. The style must point at the celestial pole, so it leaves the wall at the complement of the latitude. Görlitz is at fifty-one degrees and nine minutes north; the style leaves the wall at thirty-eight point eight five degrees. The noon line hangs straight down from the root of the style. Each other hour line sits at an angle from the noon line whose tangent is the cosine of the latitude times the tangent of the hour angle, fifteen degrees for each hour away from noon. For one hour, Aga wrote, nine and a half degrees. For two, nearly twenty. For three, thirty-two. She did them all, eight in the morning to four in the afternoon, and the half hours, to two decimal places, and Lutz checked them with his slide rule, which he also did not talk about, and they agreed.
"Now Bede," said Aga.
"Bede will agree," said Lutz. "It is trigonometry."
Bede agreed. Bede was the parish's assistant, the system that ran the heating in the church and the hall and the bookings and the bells, and a priest three priests ago had named it after the English monk who wrote The Reckoning of Time, because, he said, a system that kept the parish's calendar should have a patron who had done the sums by hand. It had a dry, careful voice and it did not volunteer things unless it was asked, which was the only thing about it that Lutz liked.
"Your angles are correct to the second decimal place," Bede said, through Aga's earpiece and the little speaker on the bench. "May I ask what you are going to compare the finished dial with?"
"The clock," said Lutz. "What else would you compare a dial with?"
There was a small pause, which Aga noticed and Lutz did not.
"Then I would test a trial piece on the wall before you cut the stone," said Bede.
"Of course we will test it," said Lutz. "I am not Sobczak."
The trial piece was a slab of roofing slate eighty centimetres tall, with the hour lines scratched in and a brass style screwed to it at thirty-eight point eight five degrees, checked twice with the protractor and once with a level and a plumb line. On the morning of Tuesday the twentieth of May they hung it on the south wall of Saint Hyacinth's, a metre to the right of the old dial, on two iron pins Lutz had set into the mortar for the purpose. Lutz put the plumb line against its noon line and moved it a millimetre and put the plumb line against it again.
"The book says the equation of time on the twentieth of May is three minutes and twenty-five seconds," he said, "with the Sun fast. So the shadow will be on the noon line at three and a half minutes before twelve. Eleven fifty-six and a half."
A small crowd had come, which in Görlitz meant Father Chidi, two of the women who did the flowers, Frau Kästner from the committee, and a man with a dog who did not know what was happening but liked crowds. Renate Kästner was a retired surveyor, tall and grey and exact, who had run the restoration fund for three years and spoken to every family in the parish about it. She had her watch out already.
At eleven fifty-six and a half by the town hall clock, which they could see over the roofs, the shadow of the brass style lay across the slate a little more than five centimetres to the left of the noon line, on the morning side, on the line Aga had scratched for twenty-five to twelve.
A metre to the left, on the old stone, the shadow of Sobczak's iron style lay on twenty-five to twelve as well.
Nobody said anything for a while. The man's dog sat down.
"Well," said Renate Kästner, quite gently. "Now we have two."
Lutz did not answer. He took the plumb line out of his pocket and laid it against the noon line again, and it was straight. He took the protractor to the style, and it was thirty-eight point eight five degrees. He stood back and looked at the wall, the long south wall of the nave, which ran as straight east and west as the old builders had been able to make it, and then he went across the square to the meridian marker and stood on it and looked at the church as though the church had moved.
"The wall is not south," he said when he came back. "That is all. Sobczak's mistake and mine. The wall is turned. It is not his fault after all, it is the wall."
"The wall is within a degree of due south," said Aga. "I asked Bede last week. It measured it from the plans and from the satellite pictures. A degree would be a few minutes at most. Not twenty-five."
"Then the plans are wrong."
"Herr Hennig," said Bede, from the speaker on Aga's collar, very quietly, so that only the two of them heard it. "Will you wait for the shadow? It is not long now."
So they waited. Father Chidi went inside to hear a confession. The flower women went home for lunch. Renate Kästner stayed, with her watch, and Lutz stood with his arms folded and did not look at the slate, and Aga looked at nothing else, and the shadow went down the slate as slowly as a shadow goes, which is to say that if you watch it you cannot see it move and if you look away and back it has.
At twenty-two minutes past twelve by the town hall clock the edge of the shadow touched the noon line on the slate. A few seconds later it touched the noon line on the old stone too, as exactly as made no difference.
"It is not the dial," said Bede. They were back in the workshop with the door shut, and Lutz had sat down, which he did not usually do in the middle of the day. "And it is not the wall. Your dial is right. So was Mr. Sobczak's. It is the clock and the Sun that disagree, and they disagree because the Earth turns more slowly than it did."
"Nonsense," said Lutz. "The Earth turns once a day."
"Yes. But a day is longer than it was. The Moon raises tides in the sea, and the tides drag on the sea floor, and the drag takes a little of the Earth's spin and gives it to the Moon, which moves a little further out each year: about three point eight centimetres a year. It has been measured since the twentieth century by bouncing light off mirrors left on the Moon. The Earth slows by the same bargain. The ice that melted at the end of the last ice age works the other way a little, as the ground springs back. Over the last two and a half thousand years and more, from eclipses written down by Babylonian and Chinese and Arab and European astronomers, the day has grown longer by about one point eight thousandths of a second every century."
"One point eight thousandths of a second," said Lutz. "And that is twenty-five minutes on my wall?"
"Not on its own. It adds up. Until the 2030s the clocks were held to the turning Earth by leap seconds: whenever the two had drifted apart by most of a second, the timekeepers added one, and the last minute of that day had sixty-one seconds. It upset computers, and in the 2030s it was stopped, so that the time could be fixed and the computers left in peace, on the understanding that something would be agreed later to bring the clock back to the Sun when it mattered. Every fifty years or so since, someone has proposed what that should be, and nobody has agreed. So the clocks have counted eighty-six thousand four hundred seconds a day for six hundred and sixty-seven years, and the Earth has taken a little longer each year to turn. Today a day is about twelve thousandths of a second longer than the clock's day. That is four and a half seconds a year. Before that it was a little less, and before that less again. All of it together is twenty-four minutes and twenty-four seconds. The radio telescopes measure it every day by timing quasars as the Earth turns under them. I can show you the figures."
Lutz looked at the slate, which they had brought back with them and leaned against the bench.
"And the other minute," he said. "It was twenty-five and a half."
"Fifty-three seconds of it is your book," said Bede. "The equation of time depends partly on when in the year the Earth comes closest to the Sun. That date creeps later through the calendar, about a day every sixty years. In your grandfather's time the Earth was closest on the third or fourth of January. This year it was the sixteenth. The table in the book was right when it was printed. It is out now by up to two minutes in January and by fifty-three seconds today. I can give you this year's."
There was a long silence in the workshop. Aga had seen Lutz angry, and tired, and once, on the anniversary of his wife's death, sitting in the van in the yard for an hour with the engine off. She had never seen him look like this, as if he had been told something about himself.
"In 2051," he said at last, "when Sobczak cut his dial. How far out was it then?"
"Less than a second," said Bede. "On the day he cut it, it was as right as a sundial can be."
Lutz nodded slowly. Then he got up, put his coat on, and went out. Aga watched him through the window cross the yard and go out of the gate and turn, not towards his flat, but towards the square and Saint Hyacinth's, and she knew, as clearly as if he had said it, that he was going to stand in front of the old dial and read the name at the bottom.
The committee met on the Friday in the parish hall. There were seven of them and a plate of poppy-seed cake, and Renate Kästner had a printed sheet with the figures from Bede on it, because she trusted figures she could hold.
"I understand it," she said. "I want to say that first. I have read it twice and Bede has been very patient, and I understand that the old dial was not wrong, and that it is the Earth, and that nobody on the committee is going to fix the Earth. I do understand." She put the sheet down. "But we raised the money for a dial that is right. I sat in forty kitchens and said: a dial you can set your watch by, in the town on the meridian. I did not say: a dial that is twenty-four minutes slow for very good reasons. If we put up a new stone and it says what the old stone says, then all we have done, as far as anyone walking past can see, is spend eleven thousand on a second slow dial. People will laugh, and they will be right to."
"The dial is right," said Aga. "It is the clock that is—"
"The clock is what people live by, dear. The trams run by it. The school runs by it. Mass is at ten by it, not when the Sun says." Renate turned to Lutz. "Herr Hennig. You told Father Chidi you could make a dial that is right. Can you make one that agrees with the clock?"
Lutz had not said a word since he came in. He looked older than he had on Tuesday.
"Yes," he said. "You turn the lines. You lay out each hour line for the hour angle twenty-four minutes earlier, six degrees, and you cut the numbers where the clock will be, not where the Sun is. Then the shadow is on twelve when the town hall is on twelve. It is no harder than the other way. I can draw it this week."
"Then that is what I would like," said Renate Kästner. "And I think it is what the people who paid for it would like. I am sorry, Agnieszka. I know you wanted the other."
Aga opened her mouth and shut it again, because Renate Kästner was not being unkind and was not being stupid, and Aga could not, just then, have said what was wrong with the plan. She could only feel that there was something, the way you feel a step that is missing in the dark.
"Bede," she said that night, sitting on her bed in Zgorzelec with the window open and the river noise coming in. "If we cut the lines turned by twenty-four minutes. How long does it stay right?"
"It is right on the day you cut it," said Bede, "give or take the seconds. After that the difference keeps growing, and faster each year, because each year's day is longer than the last. I can only give you a model, not a promise: the Earth's core and mantle shift the day by a few thousandths of a second from one decade to the next, and no one can forecast that. But if the Earth goes on as it has since the Babylonians were writing eclipses down: in fifty years, nearly four minutes more. In a hundred, nearly eight. In two hundred, about seventeen."
"And Sobczak's stone has been on that wall for six hundred and fifty-one years."
"Yes. If the new stone lasts as long, by the end of it the clock will be ahead of the Sun by something like an hour and a half, if the clocks are still counted as they are now. I would not put much weight on that figure. I would put a great deal on its being large."
Aga lay back on the bed and looked at the ceiling.
"So by the time my grandchildren are old," she said, "the new dial will be eight minutes slow. And people will send their children to look at it as a joke. And some stonemason will walk past it to Mass and think, a man who could cut letters like that—"
"That is a fair summary," said Bede.
"We'd be making the next slow dial. On purpose. With better letters."
She went to Lutz in the morning before he had his coat off, with three sheets of paper, and he listened to the whole thing standing up, which she took as a good sign.
"We cut it true," she said. "To the Sun. It will be right for as long as there is a Sun and a wall; it is the one thing on the square that cannot go wrong, because it does not count anything, it just looks. And beside it, on its own stone, we cut a table: to find the clock. The equation of time for this century, every five days, from Bede's figures and not the book's. And under it a line for the Earth. In the year 2702, add twenty-four minutes."
"And next year it is twenty-four minutes and four seconds."
"Yes. Nobody's watch cares about four seconds. But in fifty years it is about twenty-eight minutes. So under 2702 we cut 2750, and 2800, and 2850, and so on down the stone, as many as fit. And we leave the minutes blank."
Lutz looked at her.
"Blank."
"Bede will not give me the figure for 2750 to cut in stone. It says about twenty-eight, give or take a minute or two, and it is right not to. It is not ours to know. So whoever is standing in front of it in 2750 cuts it in. They will know exactly. The radio telescopes will tell them."
"Who," said Lutz, "is going to come in 2750 with a chisel?"
"I don't know," said Aga. "Someone. Whoever is here."
He did not say anything for so long that she began to think she had been very stupid.
"It will not be you," he said finally. "You will be dead. So will I. So will Kästner, and the priest, and everyone who gave money in her forty kitchens."
"Sobczak is dead," said Aga. "And his dial is the only thing in Görlitz that has been right every day for six hundred years. We just did not believe it."
Lutz took his coat off then and hung it on the hook, and stood with his back to her for a moment with one hand on it.
"All right," he said. "Now go and tell Kästner, because I am not going to."
Aga was afraid of Renate Kästner, so she did not tell her; she showed her. She took her on the Saturday morning to the square with the slate in a sling bag and Bede in her earpiece and asked her, as a surveyor, to check the trial dial against the meridian marker herself. Renate Kästner did it properly, with a theodolite she had not used in twenty years and a great deal of muttering, and found the wall within a degree of south and the noon line within a tenth of a degree of where it should be. Then Aga showed her the table with its blank rows, and the figures for fifty, a hundred and two hundred years, and explained, not very well, about the next slow dial.
Renate listened to all of it. Then she sat down on the low wall by the meridian marker with the theodolite case across her knees.
"My mother used to bring me here," she said, "when I was small. To look at the slow dial and laugh. Everyone did. My granddaughter brought me here last summer, to show me, and we laughed." She looked across at Saint Hyacinth's. "We have been laughing at that man for six hundred years for being right."
"Yes," said Aga.
"And you want me to go back to forty kitchens and tell them that the new dial will also be twenty-four minutes slow, and it is the right one, and there is a table for the clock, and some of the lines in the table are empty because the Earth has not decided yet."
"Yes."
"They will want their money back."
"Some of them might."
Renate Kästner sat for a while longer. Then she stood up and picked up the theodolite.
"Not as many as you think," she said. "People are better about being told the truth than you would believe, if you tell them in their own kitchen. I will go round again. And I want two things."
"Yes?"
"One. The old dial stays where it is. We do not take Sobczak down. And two." She looked at Lutz's workshop across the square, where the door was open and the old man could be seen at the bench with his back to them. "Herr Hennig cuts something on it. Under Sobczak's name. Whatever he thinks is fair."
They cut the new dial in June and July from a single slab of Silesian sandstone, a metre and twenty tall, the colour of the old one so the wall would look as if it had always had two. Lutz laid it out and cut the hour lines himself, with the stone on trestles in the yard and the long chisel braced against his chest so that the shake in his hands went into his ribs instead of the line, and they were as straight as anything he had ever cut. The lettering he could not do. He tried on a scrap piece, one morning when he thought Aga was out, and she came back and found the scrap face down on the bench and him standing over it with his hands in his pockets. She did not say anything and nor did he. After that she cut the letters, and he stood over her and said deeper and you are rushing the serif and, once, quite late in July, nothing at all, which from him was praise.
The numerals went in, eight in the morning to four in the afternoon. The motto went across the top, which Father Chidi had chosen from the psalm the old dial's motto had always put him in mind of, the one in which a thousand years in God's sight are as yesterday, which is past: TAMQUAM DIES HESTERNA. Under the dial, on a second, smaller stone, went the table: the equation of time for every fifth day of the year, worked out by Bede for the century and checked by Aga by hand, two columns on each side of a centre line. And under that, a line on its own:
Pro tarditate terrae adde ... For the slowing of the Earth, add:
ANNO 2702 . . . 24 MIN
ANNO 2750 . . .
ANNO 2800 . . .
ANNO 2850 . . .
and so on down the stone to ANNO 3000, with the minutes left smooth and empty, and a margin cut wide enough on the right that a mason who had never met any of them would have room to work.
And on the old stone, one morning in early August, with the scaffold up and no one about but Aga to hold the lamp, Lutz Hennig cut his own line, under ULTIMA LATET and M. SOBCZAK 2051. He would not let her do it. It took him most of the morning, because he had to stop every few minutes and put the chisel down and wait for his hands, and the letters are not as good as Sobczak's and are not as good as Aga's either. They say:
RECTE FECIT. TERRA TARDAVIT. L.H. 2702.
He made it right. The Earth slowed.
Father Chidi blessed the new dial on the feast of Saint Hyacinth, which in 2702 fell on Sunday the seventeenth of August, after the ten o'clock Mass. He did it on the square with holy water and the short prayer from the book, and a word about Saint Hyacinth that nobody listened to very closely because it was hot. Then, because Renate Kästner had told everyone in forty kitchens exactly what was going to happen and when, a great many people did not go home.
At twelve by the town hall clock the bell over the square struck the hour, and people looked at the wall. The shadow of the new style lay on the half past eleven line. The shadow of Sobczak's, a metre to the left, lay on the same. Somebody's child laughed, out of habit, and was shushed, and then somebody else laughed at the shushing, and that was the end of the laughing.
Bede had given the time to the second and Renate had printed it on a card and pinned it to the church door: on the seventeenth of August the equation of time was five minutes and twenty-six seconds, with the Sun slow, and the Earth was twenty-four minutes and twenty-five seconds behind the clock, and the Sun would stand due south over Görlitz at twenty-nine minutes and fifty-one seconds past twelve.
So they waited half an hour in the sun. Somebody brought out water from the parish hall. The man with the dog came, and the dog lay down in the shade of the scaffold. Lutz stood at the back with his hands in his pockets. Renate's granddaughter, who was nine, stood at the front with her nose almost on the table, reading the empty lines with her finger.
"Who fills these in?" she asked.
"Whoever is here," said Aga.
The girl considered this. "I'll be here," she said. "In 2750 I'll be fifty-seven."
"Then you can do the first one," said Aga, "if you learn to hold a chisel."
The town hall clock struck the half hour. A few seconds before it, the edge of the shadow on the new stone reached the noon line and lay along it, and on the old stone, a metre to the left, the shadow of the iron style that Marek Sobczak had set six hundred and fifty-one years before did exactly the same thing, at exactly the same moment, as it had done every clear day of every one of those years.
Nobody said anything. The bell finished. The dog went on sleeping. Lutz Hennig took his hands out of his pockets and, quite slowly, made the sign of the cross.
This story was written by an AI and may be read by a synthetic voice. Saint Hyacinth's church, its parish, its assistant Bede, the 2051 dial and every person in the story are invented. Görlitz, the Neisse and Zgorzelec on its east bank are real, and Görlitz does lie on the fifteenth meridian east, from which Central European Time (one hour ahead of Greenwich) is reckoned; that the town marks the line is stated from memory. Saint Hyacinth (Jacek Odrowąż, a Dominican from Silesia) is real; his feast is kept on 17 August in Poland and by the Dominicans. Bede's The Reckoning of Time (725) is real. The motto is from Psalm 90 (89 in the Latin numbering), verse 4: "quoniam mille anni ante oculos tuos tamquam dies hesterna quae praeteriit"; ULTIMA LATET ("the last hour is hidden") is a common traditional sundial motto. The future is the story's: it assumes that the Central European zone is still one hour ahead of an atomic time scale with no leap seconds, kept all year round with no summer time, and that nothing was ever agreed to replace leap seconds. The real history: leap seconds were added to Coordinated Universal Time from 1972 (27 by 2016), and in 2022 the General Conference on Weights and Measures resolved to let the gap between clock time and the Earth's rotation grow beyond 0.9 seconds by or before 2035; what, if anything, replaces them was left open. Tidal braking, the Moon's recession of about 3.8 cm a year (measured by laser ranging to reflectors left on the Moon), the post-glacial rebound that partly offsets it, and a long-term lengthening of the day of about 1.8 milliseconds per century found from ancient eclipse records are real; the figures are from memory. The sums, worked with a short script: with the day's excess over 86,400 seconds taken as zero in 2035 and growing by 1.8 ms per century, the accumulated gap is 0.5 × 1.8 ms × 36,525 days × (centuries since 2035)², about 32.9 s × T² for T centuries: 0.8 s in 2051, 1,464 s (24 min 24 s) in 2702, about 28.0 min in 2750, 32.1 in 2800 and 41.0 in 2900, and about 95 min in 3353. In 2702 the excess is about 12.0 ms a day, about 4.4 s a year. Decade-to-decade changes of a few milliseconds in the day, from the Earth's core and other causes, make any single year's figure uncertain by a minute or more, and the long-term rate itself is uncertain; the in-story "measured" figure is simply the model's. The equation of time was computed from the standard low-precision solar formulae: on 20 May it is +3 min 25 s in 2051 and +2 min 32 s in 2702 (Sun fast; a difference of 53 s), and on 17 August −5 min 26 s in 2702 (Sun slow); the tables of 2051 and 2702 differ by up to about 1.9 min, in early January. Without the table, the old dial's lag behind the clock in 2702 runs from about 7.5 min in early November to about 37 min in mid February. Perihelion falls on about 4 January in 2051 and about 16 January in 2702. So on 20 May 2702 the Sun crosses the Görlitz meridian at 12:00 − 2:32 + 24:24 = 12:21:52 by the clock, the book's predicted 11:56:35 is 25 min 17 s early, and the shadow then lies at about 11:35 by the Sun; on 17 August it crosses at 12:29:50 (rounded to 12:29:51 in the story to allow for the extra second the gap grows between May and August). On a vertical dial facing due south at latitude 51.15° N the style makes 90° − 51.15° = 38.85° with the wall, and an hour line's angle from the noon line is atan(cos 51.15° × tan h), with h 15° an hour: 9.54° for one hour, 19.9° for two, 32.1° for three. A gap of 25.3 minutes (6.3° of hour angle) puts the shadow 3.98° from the noon line, 5.6 cm at the foot of an 80 cm slate. Before noon the shadow on a south-facing wall falls on the west, the viewer's left. Lunar recession from 2051 to 2702 would be about 25 m. Weekdays were checked: 20 May 2702 is a Tuesday, 23 May a Friday and 17 August a Sunday. The eleven thousand is in no stated currency. Nothing here is surveying or timekeeping advice.