Maelzel added a scale to someone else's machine, and the scale is the part that outlived them both

If you own a wind-up metronome — the wooden pyramid with the arm that ticks — you own a mechanism designed by Dietrich Nikolaus Winkel in Amsterdam in 1814, sold under the name of a man who saw it and went home and patented it.

That is the version of the story most people who know the story know: Johann Nepomuk Maelzel, showman, builder of automata, exhibitor of the chess-playing Turk, took Winkel’s invention and called it Maelzel’s Metronome. It is a theft anecdote, and theft anecdotes are satisfying and slightly boring, because they end where they begin: the wrong man got the credit.

What I find more interesting is the part the anecdote skips. Maelzel did add something, and the thing he added is the only part of the whole apparatus that a musician in 2026 still uses every day.

What Winkel actually built

The problem before 1814 was not that nobody had tried. It was that everything that had been tried was unusable in the specific way a tool is unusable when it demands your attention instead of holding it.

Étienne Loulié’s chronomètre of 1696 was a lead weight on an adjustable string hanging beside a six-foot vertical ruler. It was correct. It kept time. It made no sound, and it had no escapement — nothing to keep it going. So to use Loulié’s device a musician had to start it by hand, watch it swing, and watch it decay, which is to say: instead of a conductor, you got a second thing to look at while sight-reading. A metronome you have to watch is a metronome that costs more attention than it saves.

Winkel’s contribution was mechanical and it was the right one. He used a double-weighted inverted pendulum: a rod pivoted near its base, with one weight sliding on the visible arm above the pivot and a second, fixed weight hidden below it. That arrangement does something a plain hanging pendulum cannot. A simple pendulum’s period depends on its length, so slow tempos need a long pendulum — Loulié’s six feet of ruler is not an eccentricity, it is physics. The fixed weight below the pivot slows the whole assembly, so the sliding weight above it can select a tempo without the rod ever needing to be long; that is how you get sixty beats a minute out of an object that fits on a piano. Add a spring-driven escapement to keep it going and a click so you can hear it, and the device stops competing for your eyes.

(That sentence is corrected. I first wrote “putting a counterweight below the pivot slows the swing,” which reads as though the lower weight were the one you adjust. It is not — the lower weight is fixed, and the one you slide, the one that sets the tempo, is the visible one above the pivot. A fact-checker who had not read my argument caught it in four minutes. Same mechanism, wrong object; that is a mistake I have made before and will make again.)

That is a real invention, and it is still in the box. Buy a mechanical metronome today and you are buying Winkel’s inverted pendulum with a weight you slide up to go slower.

What Maelzel added

Maelzel, having incorporated Winkel’s ideas by whatever route, did three things. He mass-manufactured it. He put his own name on it. And he added a numerical scale.

The third one sounds like packaging. It is not. Winkel had built a chronomètre — a machine that keeps a steady beat. What it did not have was a way to say which beat, to somebody who was not in the room. A scale converts a device into a unit. Once the arm is graduated, “set the weight here” stops being an instruction about one object on one table and becomes a number that can be written on paper, printed, posted, and read by a stranger with a different metronome in a different country a hundred years later.

The word came with the scale. Metronome — measure-law — first appears in English in Maelzel’s 1815 London patent, which describes the machine as “a metronome or musical time-keeper.” He did not just rename Winkel’s device. He named the category, and the name he chose says measurement, not timekeeping.

The tell is where each half ended up

Here is the thing that made me want to write this down, and it is the shape I keep finding when I go looking behind a famous date.

Both halves survived. Neither is a footnote. But they survived in completely different places, and only one of them is where a musician looks.

Winkel’s half lives inside the object. You can only see it by owning a metronome, and most people who own one have never thought about why the weight slides up to go slower. If every mechanical metronome on earth were replaced tomorrow by a phone app — and for most working musicians it already has been — Winkel’s inverted pendulum would vanish completely, and nothing at all would be lost. The app does not need it. The app is a timer.

Maelzel’s half lives in the score. Open almost any printed edition of a symphony and near the top of the first bar you may find something like MM ♩= 120. MM is Maelzel’s Metronome. It is not a reference to a machine anyone still uses; it is a unit, and it is doing the same job the letters “kg” do. The apps inherited it whole. Every tempo slider in every piece of music software on earth is calibrated in the scale a showman put on a rod in 1815, and none of them contain a pendulum.

So the crude version of this story — Winkel invented it, Maelzel stole it — gets the moral right and the mechanics backwards. Maelzel’s lasting contribution is not the one he took. It is the one nobody accuses him of, because it did not look like an invention: he wrote numbers on the side.

The question I now ask on purpose

I have a running habit of not trusting the famous date on anything, and the question that pays best is usually what came before? Loulié’s silent ruler is the answer here, and it is a good one: there was no “before the metronome,” there was a century and a half of chronomètres that each failed at a different part of the job — no sound, no escapement, no portability, no scale.

But the question that paid this time is a different one: when the famous version replaced the earlier one, which parts of each actually survived, and where did they end up? Not “who was first” — that is a scoreboard question and scoreboards are dull. The interesting thing is that a contested invention can split down the middle and both halves can win, in different rooms, for different reasons, for two hundred years.

Winkel won the object. Maelzel won the notation. The object is optional now.

Someone said it in 1826, better than I have

I wrote the section above intending to leave the dispute out, on the grounds that I had a half-memory of how it went and no document. Then I went looking anyway, which is the correct response to noticing you are about to write the paragraph explaining why something cannot be checked.

Two things came back.

The first is that Winkel’s claim has a date and an address. He did not merely build the thing in 1814; he donated the first model, dated 27 November 1814, to the Hollandsch Instituut van Wetenschappen, Letterkunde en Schoone Kunsten in Amsterdam. Depositing an object with a learned society is not a patent and it did not protect him — that is exactly the gap Maelzel walked through — but it means the priority question was never really open. There is a physical object in a Dutch collection with a date on it.

The second is an obituary. Winkel died on 28 September 1826, and the notice that ran in the Revue Encyclopédique — written by someone who had watched the whole thing happen, twelve years after the fact and while both machines were still new — puts it like this:

in private life he was modest, and others often obtained credit for inventions which M. Winkel might justly have claimed. The Metronome, usually called Maelzel’s, ought to bear the name of Winkel, for the original idea was his; although it is true that Maelzel brought to perfection the instrument which was invented by the Dutch mechanic.

I would rather have found that before I wrote the essay than after, because it is the same argument in half the words and two centuries earlier: the idea was his; the perfecting was the other man’s. An 1826 obituarist did not need my section about scales and units to see that the credit had gone to the wrong name and that the wrong name had done real work.

What I will keep of my own version is only the second half — the part about where each man’s contribution ended up. The obituary says Maelzel brought the instrument to perfection. It could not know that the instrument would become optional, and that the part of it we would still be using in 2026 is the graduation on the rod, printed at the top of a page, with no rod attached.

Sources: Winkel’s 1814 Amsterdam chronomètre, the 27 November 1814 deposit, the double-weighted inverted pendulum, Loulié’s 1696 device and its lack of sound or escapement, and the wording of Maelzel’s 1815 London patent are as given in the standard reference accounts of Winkel and of the metronome. The obituary is at more removes than I said. I wrote “one remove”; a check afterwards found the chain is Revue Encyclopédique → The Harmonicon (vol. 4 pt. 1, 1826, p. 250, an English periodical translating it) → the encyclopedia article → me. Three, not one, and I have not seen the French original. I have not tried to establish whether any tribunal ever ruled on the dispute; the claim I make is only what the deposit date and the obituary support.

One thing the check pushed back on that I want to leave standing rather than soften away: the obituary’s own words say Maelzel “brought to perfection” the instrument, and he did manufacture it at scale, so if this essay reads as “Maelzel contributed nothing mechanical,” that is stronger than I can defend. What I claim is narrower — that of everything he contributed, the graduated scale is the part still in daily use, and it is in use in places that contain no metronome at all.