How Clean Is Too Clean

The Sistine Chapel ceiling was cleaned between June 1980 and April 1994, under Gianluigi Colalucci and the Vatican Museums staff. Fourteen years.

What came off was five centuries of accumulation: candle smoke, airborne dust, and the glue and resin that earlier restorers had applied in attempts to revive a darkening surface. Those earlier materials had themselves darkened, so part of what was being removed was previous good intentions.

What emerged underneath was colour. Not slightly brighter: a palette of pinks, acid greens and oranges that nobody alive had seen and that contradicted several generations of writing about Michelangelo's sombre grandeur. Art historians had built an account of his temperament partly out of what turned out to be soot.

Then the objection, and it is worth taking seriously because it is technically specific rather than a general complaint about meddling.

Fresco painters of the period often finished a passage a secco: after the plaster had set, they went back in with paint bound in a medium that sits on the surface rather than becoming part of the wall. Adjustments, deepened shadows, final detail. Those layers are not locked into the lime. They are, chemically speaking, in the same position as the dirt.

James Beck, an art historian at Columbia, began pressing this case publicly in 1986. If Michelangelo had made his final decisions in secco, then a cleaning aggressive enough to take off five hundred years of smoke would take those decisions off too, and the bright ceiling everybody was admiring would be an unfinished one. He called it an artistic Chernobyl. In 1987 a group of artists including Christo and Andy Warhol petitioned John Paul II for a precautionary pause.

The restoration continued. Reviewers who went through Beck's specific claims reached a mixed verdict: some objections were subjective, some rested on an incorrect reading of the historical documents, and some were reasonable questions that needed answers.

Here is the structural point, and it is the reason this chapter exists rather than a summary of who won.

The dispute cannot be settled by examining the ceiling. If the secco layers were removed, the evidence that they existed went with them. Nothing you can look at today distinguishes a ceiling that never had final dry touches from a ceiling that had them and lost them. The only materials that could support either position are the photographs, the analyses and the records made before and during the work, which is precisely why the previous lesson is about documentation rather than about solvents.

The Last Supper restoration is the mirror image and makes the same point from the other side. Twenty-one years of work under Pinin Brambilla Barcilon, finishing in 1999, on a wall where Leonardo's experimental technique had been failing since the sixteenth century and very little original paint survives. The complaint there was not that too much had been removed. It was that too much had been reconstructed: that the visitor is looking at a modern painting with Leonardo's composition.

Removal and reconstruction, opposite accusations, one underlying question. How much is a restorer allowed to decide, and how would anybody afterwards know what they decided.