Monet painted Rouen Cathedral thirty times. The cathedral wasn't the subject.
In the winter of 1892 Monet rented a room across from Rouen Cathedral and painted the same stone under thirty lights. It was the most disciplined color study in painting history. How to run the exercise in ten canvases.
In February 1892, Claude Monet rented a room on the second floor of a dress shop on the Place de la Cathédrale in Rouen, directly across from the west facade. He set up as many easels as the room could hold, sometimes nine at once, and for the next two winters he moved between them through the day. When the light shifted, he abandoned one canvas and picked up the next. The same thirty meters of Gothic stone, the same angle, through fog, through flat morning, through full sun, through blue twilight, through gold late sun, through rain.
By the end he had thirty finished paintings. The cathedral is the same in every one. Nothing else is.
This is the clearest demonstration in painting history of what a color study actually is. It is also the clearest explanation of why digital illustrators, technically better equipped than Monet in every respect, learn color more slowly than he did.
What Monet was actually doing
The usual reading of the Rouen series is that Monet was interested in the cathedral, or in Gothic architecture, or in the French church. None of that is correct. He wasn’t particularly religious. He’d already painted anything that held still long enough: haystacks, poplars, the Gare Saint-Lazare, his own garden.
What those subjects shared wasn’t that they were inherently interesting. It was that they could serve as a holder. A constant around which everything else could vary.
The thirty Rouen canvases aren’t thirty paintings of a building. They’re thirty paintings of light on a building. The cathedral is the control variable. The light is the independent one. Thirty different subjects under thirty different lights would have taught him nothing. What you can’t hold steady, you can’t compare.
He wrote to Alice Hoschedé, his partner and later wife, from that rented room, about being exhausted, about nights of nightmares in which the cathedral kept falling on him, alternately blue, pink, and yellow. That isn’t a nightmare a normal person has. It’s the nightmare of someone who has spent eight hours watching blue stone turn pink at four in the afternoon.
Why illustrators learn color slowly
Here’s the mistake, and almost everyone reading this has made it: you paint thirty different pieces this year and think you’ve done thirty color studies. You haven’t. You’ve painted thirty pieces that had color in them. It’s a completely different exercise.
A painting is a composition, a subject, a set of shapes, and a color decision, all at once. When the result doesn’t land, you don’t know which variable failed. Maybe the composition. Maybe the values. Maybe the color. Probably some combination. You can’t isolate what you learned, which means you didn’t really learn much, which is why the piece you finished last week and the piece you finished a year ago overlap far more than they should for someone practicing every day.
A color study isolates color by locking everything else. Same subject, same composition, same time invested. Only the color changes. Now when the piece works or doesn’t, you know why, because nothing else was different.
Monet built himself a one-person color laboratory in Rouen. Most illustrators never do this, and then wonder why their color isn’t improving.
The ten-canvas version
You don’t need a rented room in Rouen. You need one iPad canvas and the willingness to repeat yourself.
Pick a motif. Not a portrait, not a character, not a scene with story in it. Something simple enough to draw in twenty minutes: a still life of three objects, a single tree, a window, a roof against a sky. The motif should be small enough that redrawing it a second time annoys you. That annoyance is the point.
Copy the motif ten times across ten canvases. Same composition, same scale, same underdrawing. Canvas one and canvas ten start identical.
Then paint each one under a different lighting condition. Morning haze, overcast, golden hour, blue hour, warm indoor tungsten, cold fluorescent, firelight, vertical noon, storm light, moonlight. Ten conditions you can name. One condition per canvas. Different palette every time. The motif does not move.
This is the exercise Monet was doing. You’re not copying him. You’re doing the same kind of thing, at the smaller scale your life allows.
What you notice by study seven
The first three feel useless. You fight your defaults, you overcorrect, the green you pick for the overcast version is still contaminated by the sunny one. Keep going.
Somewhere around study seven or eight, something shifts. You stop picking colors by noun (“the sky,” “the tree”) and start picking them by condition (“cool top-down light falling through grey air”). The names for your colors get longer and more specific. You start deciding based on how the light acts on the motif, not what the motif is.
That shift is the entire reason the exercise exists. The name for what you’ve gained is color seeing, and before study seven you didn’t have it, and after study ten you can’t turn it off.
Once you see that way in ten controlled studies, you start seeing it in the uncontrolled world. You look at a reference photo and notice the light condition before the subject. You look at a finished piece of your own and know immediately whether the light is consistent across the image or whether you picked colors by noun and the piece is a collage of four different times of day.
What it takes
A calendar entry: “color study, 7pm, any motif.” An agreement with yourself to do one every other day for a month. A named palette library where you save the five-to-eight colors per study with the condition as the title. That’s the whole setup. The studies don’t have to be good paintings. They have to be ten different answers to the same question.
Monet exhibited the thirty cathedrals together because their value was collective, not individual. Each painting is interesting. The series is important. Your ten studies work the same way. No single one of them is a piece. Viewed as a grid, they’re the best evidence you’ll ever produce that your color has a voice.
If tonight you don’t know what to paint and you’re about to start another one-off, don’t. Pull up yesterday’s motif, change the light, paint it again. That one, set next to yesterday’s, is worth more than either of them alone.