Mystery of blue
The color theory greatly comes into play in retouching and repainting. The process involves mixes and matches made through experience, gut instinct, and trial and error on a small palette - it's the only way to solve the Pantone puzzle. As Walsh reflects, blue is perhaps the most enigmatic color. Coincidentally, blue is also one of the most frequently recurring and iconic shades among ancient Chinese artifacts. Consider the sinuous five-clawed dragon, inked in blue, coiling around the neck and body of a teapot from the Ming Dynasty (1368-1644), exuding vigor and power; or the blue cloud scrolls - xiangyun in Chinese - that adorn Chinese art, architecture, furniture and textiles for their auspicious meanings; or the interlocking swirling peony flowers, rendered in cobalt blue, that take pride of place on Chinese porcelain vases, symbolizing health, prosperity and harmony.
"The issue with blue is metamerism which means that blue will look different under different lighting. If you look at a blue object outside in sunlight, there's a very good chance you'll see a nuanced blue under artificial lighting indoors. The blue changes," Walsh explains.
What should a Chinese art restorer aim to achieve? "A blue that fits all environments, indoors and outdoors," she says, while displaying a rack of paints shaped like drip tubes used in laboratories, each one labeled with its precise name - ultramarine blue, cobalt blue, cerulean blue. "In blue-and-white Chinese ceramics, even within a single object, there can be a dazzling array of shades. It's more complicated than it seems. You build it up in layers with thin coats of paint before you can try to replicate the tonal variations."
She then lays four color tubes on the table, as if presenting a verdict on the formula for "traditional Chinese porcelain" - flake white, lemon yellow, raw umber and ultramarine. "You see, Chinese porcelain is never purely white. It always has a very faint greenish blue tint," says Walsh, revealing her discovery on a porcelain artifact.