Paint Schoodic

Join us in Acadia National Park in August. Click here for more information.

Wednesday, March 28, 2018

Mixing complements and making grey


Some people say it doesn’t work. Is that true?

All flesh is as grass, by Carol L. Douglas. Since the Impressionists we have mixed our grays with complements.
Painters use mixes of complementary colors to make neutrals: red and green, blue and orange, or yellow and purple. The exact mixes have to be juggled around depending on the paint, but it’s an efficient system to get soft greys and browns. It’s centuries old and it endures because it’s a useful system.

Yesterday, a student flummoxed me by asking why it works. I could answer in general terms—interference—but I really didn’t know in any detail. I started to read about it and came up with a striking problem: many people don’t believe it actually does work.

From The Natural System of Colours, 1776, by Moses Harris. Courtesy Project Gutenberg. 
The traditional color wheel is a concept that we’ve been tinkering around with since Sir Isaac Newton and his experiments with light in the 17th century. By the time the Impressionists started their world-changing experiments with light and color, the color wheel was settled in the format we currently use: a triad of so-called primary colors (red, blue and yellow) with secondary colors inserted between them.

A complementary color pair is made up of a primary color and the secondary color that sits across from it on the wheel. For example, yellow is a primary color, and purple is made by mixing red and blue. When yellow and purple paint are mixed, all three primary colors are present.

L'air du soir, c.1893, Henri-Edmond Cross, courtesy Musée d'Orsay. Pointillism works because the eye averages adjacent spots of color into mixes.
Paints are what we call subtractive color. That means they absorb light. What we see is what’s allowed to bounce back to our eyes. Neutrals happen when no particular color bouncing back to us is able to dominate; the three primary colors cancel each other out.

So why do some scientists and artists say this system doesn’t work? Mostly, it has to do with the impurity of pigment. Historically, all pigments were approximations of pure color, based on what technology could produce.

Our paints never sit exactly on the point of a primary or secondary color. Furthermore, there are a million sets of complements. For this reason, I devised a class exercise based on Stephen Quiller’s painter-specific color wheel, so that my students could find beautiful combinations based on the pigments they actually use. If you missed this lesson, I encourage you to try it now.

Fish Beach, by Carol L. Douglas.
Traditional pigments also change with concentration. We’ve all experienced this: three different reds may look the same out of the tube but end up looking very different when diluted or mixed with white. These imperfections allow us to mix some odd combinations that shouldn’t be possible—ultramarine, which is a violet-blue, can still make a passable green. This is also why we can mix ultramarine and burnt sienna—both on the red side—and get wonderful greys. There are undertones to those pigments that gain prominence when we start manipulating them.

Twentieth century pigments were designed with industrial and commercial applications in mind. They don’t change color with concentration, so mixing historic and new pigments together sometimes yields surprising results.

It's about time for you to consider your summer workshop plans. Join me on the American Eagle, at Acadia National Park, at Rye Art Center, or at Genesee Valley this summer.

No comments: