A wall painting rarely announces that it needs attention. It goes dull, then grey, then quietly disappears under candle soot, dust, and the salts that migrate up through the masonry. Eventually someone asks for an estimate, the scaffold goes up, and the conversation jumps straight to methods and materials. The decision that matters most has already been made by then, and it is not really a technical one. It is a judgement about how far to go.
Ethics in fresco conservation is less about lofty principles than about a hundred small choices made on a cold morning with a swab in your hand: does this stain come off, should it, and what is underneath it if it does? The following is a working guide to that reasoning — the order of operations, the traps, and the questions worth asking before anything touches the plaster.
Start with the order of operations, not the method
Most bad interventions are not caused by the wrong product. They are caused by doing the right treatment in the wrong sequence, or doing a cosmetic treatment when a structural one was needed. Stabilization cannot wait. Cleaning usually can. Aesthetic choices can almost always wait.
- Understand the building before the painting. Roof leaks, blocked gutters, rising damp, a new heating system, a bus route outside the apse — the wall reacts to all of it. Treating the surface while the source of moisture continues is expensive theatre.
- Document the state of things. Raking-light photographs, a drawn or digital condition map, notes on previous interventions. You cannot prove you did no harm if you have no record of what you found.
- Stabilize what is failing. Detached plaster, powdering paint, open cracks, active salt crystallization. This is the work that protects the object and buys decades of options.
- Clean only what the paint can survive. Never as a reflex, always after tests.
- Retouch gaps, not history. If at all.
- Plan the maintenance. A conservation report that ends with the scaffold coming down is an unfinished report.
Cleaning is the one step you cannot undo
Consolidation can be repeated. Retouching can be removed with a solvent and a little patience. Cleaning removes material permanently, and the material at the surface is the part of the fresco the painter was most careful with.
That surface carries more than dirt. It carries the a secco finishing layers — azurite skies, copper greens, gilded haloes, the fine dark outlines that give a face its expression — applied after the plaster dried and therefore far more sensitive to water and alkaline cleaners than the fresco proper. It carries the patina: the aged, slightly translucent surface that saturates colour and makes a fourteenth-century wall read as a painting rather than a diagram. And it carries evidence: traces of earlier restorations, of a change of mind in the workshop, of a lost sinopia.
Over-cleaning is rarely dramatic. It looks like a wall that has become a little chalky, a little pink where it should be warm, a sky that has lost its depth. It happens when the test patch is judged in the wrong light, when the poultice is left on twenty minutes too long, when "just a bit more" becomes the working method. Practical safeguards:
- Run cleaning tests in a discreet area — near a cornice, in a shadowed corner — and leave them visible for a week before committing.
- Photograph every test in the same light, from the same angle, with a colour reference in frame. Memory is a poor witness.
- Work from the least aggressive method outward: dry brushing, soft rubber sponges, deionized water compresses, chelating or mildly alkaline poultices, laser where appropriate — and stop at the first sign that the surface is changing character.
- Check a cross-section or a magnified view before assuming a brown layer is dirt. It may be a varnish, a glue, a previous retouching, or the painter's own toning.
- Agree in writing where the cleaning stops. The best cleaning leaves a whisper of patina, not a bleached wall.
The argument about how far to clean is old and respectable: the cleaning of Masaccio's Brancacci Chapel in Florence is still discussed by conservators decades later, and not because the work was careless. Reasonable people disagreed about what counted as original surface. Expect the same argument on your own scaffold, and be able to explain your reasoning.
Stabilization: the work that makes everything else possible
Stabilization is unglamorous and usually invisible, which is exactly why it deserves the first line of the budget. The main enemies of a fresco are water, soluble salts, and movement.
Salts and moisture
Salts crystallize just beneath the paint film and push it off in flakes. Extraction with poultices can reduce the immediate problem, but it is a holding action: if the moisture source is not addressed, the salts return. Find the source first — a leaking downpipe, a raised external ground level, cement render trapping water in the wall, a blocked drainage channel in the floor.
Detachment and cracks
Where plaster has separated from the masonry, the void is filled by injection through small, carefully placed holes, with the edges supported until the consolidant sets. Cracks are stitched or grouted depending on movement. The temptation here is to cosmetically fill a crack and paint over it. Resist until you know whether the crack is dormant.
Detachment from the wall itself
Strappo, stacco, and stacco a massello are legitimate techniques with a real history, and they have saved countless paintings from dams, earthquakes, and demolition. They are also irreversible, and they take a wall painting out of the architecture it was made for. In most current practice, full detachment is a last resort, justified by genuine necessity rather than convenience.
Retouching: what a gap should look like
The purpose of retouching is not to make a damaged painting look undamaged. It is to reduce the visual noise of loss so the surviving original can be read, without confusing the eye about what is original.
Italian practice offers useful models. Tratteggio — fine parallel strokes of pure colour drawn across a lacuna — reintegrates the image at a distance while remaining unmistakable up close. A neutral toned plaster achieves something similar with less intervention. Where a large area of figure is lost, the honest answer is often to leave it quietly empty. Rebuilding a face or a hand from imagination is not restoration; it is invention, and it will mislead every viewer who comes after you.
Retouching should sit on an isolating layer, use materials that age predictably, and be removable without disturbing the original. Where a documented source exists — an old photograph, a copy, the sinopia beneath — it can guide an integration, but it should be declared in the report, not quietly absorbed into the composition.
When the pressure is not conservation
Decisions get bent by deadlines, by donor expectations, by a desire for a bright wall behind the altar, by a photograph that will appear in a magazine next month. None of these are illegitimate needs, and a conservator who ignores them will not be invited back. But the painting cannot speak for itself in that meeting.
What protects you is a written condition report, a clearly stated scope, honest options with different costs and consequences, and a note in the file when a decision is taken against advice. Saying "we can clean this much now and revisit it in ten years" is usually more useful than refusing outright — and it keeps the door open for the next generation of conservators, who will have better methods and better diagnostics than we do.
A practical checklist before you touch the wall
- Has the building's water and drainage problem been addressed, or at least understood?
- Do I know what is on the surface — dirt, varnish, glue, old retouching, secco finishing?
- Have I run tests, waited, and photographed them properly?
- Can I write down exactly where cleaning stops, and can I defend it?
- Is every material I plan to use reversible, or at least retreatable, in this specific situation?
- Have I documented the before state in adequate detail for someone who will never meet me?
- Does the owner understand that a stable, partly legible painting is a good outcome?
- Who is coming back in five years to look at it, and will they have my notes?
Pigments and solvents deserve respect in the workshop too: lead white, vermilion, and many consolidants require gloves, ventilation, and proper disposal, and local safety regulations should be followed as a matter of course.
The most ethical intervention is usually the smallest one that solves the actual problem, done slowly, documented thoroughly, and explained in plain language to whoever is paying for it. A fresco that has survived six centuries does not need to be made new. It needs to be kept readable, kept dry, and left with enough of its own surface for the next conservator to work with.
Photo: Tahir Xəlfəquliyev / Pexels
