Why is it a Color “Wheel” and Not a Color “Line”?

This article has been moved to Nature, Brain, Technology.



4 thoughts on “Why is it a Color “Wheel” and Not a Color “Line”?

  1. Dr. David Burton, Professor of Art Education at Virginia Commonwealth University, tells me that he’s been lobbying for a variety of color systems to be taught in schools, to reflect the diversity of influences, practices, and historical contexts for color theory. Regarding my blog post on color wheels, He provided the following extra background information:

    Legend has it that Isaac Newton created the first color wheel. As you recall, he did the classic experiment of breaking white light into 7 colors (ROY G BIV–Red, orange, yellow, green, blue, indigo, violet; he really only saw 6) by placing a prism in front of a beam of light coming through a slit in his curtain. Having just a slit of light is essential. If the beam is any wider, overlapping spectra recombine into white light again. Anyway, you are right, violet/red-purple, does not appear in the physical spectrum because red and indigo (purple) are at opposite ends of the spectra.

    Why then, violet? Descartes had arranged the 7 notes of the musical diatonic scale into a disc/wheel, probably to demonstrate how octaves follow one another. Newton mimicked Descartes circular format, joining the two ends of the spectrum with a credible hybrid, violet. (Of course, he didn’t see this in his visual spectrum, nor do we, but we have nevertheless labeled it “ultraviolet.”) To his credit, Newton gave each of the color as much proportional room on the wheel as they have in a spectrum–yellow is just a sliver, red is much wider. Also, remember Newton was a numerologist and 7 is a sacred number. Newton published all this in his Opticks (1704). The experiments has been done 20 years before.

    Curiously, most color wheels have an even number of colors–6, 8, or 12. This is so complements can sit opposite each other, red across from green. Of course, blue-green is the true complement of red, but that is another story related to the history of pigments. The oldest color wheel I know of is attributed to Anthansius Kirchner (1671, before Newton). It had 5 primaries and 10 secondary colors. Kirschner was also a numerologist and his color wheel has more to do with astrology and numerology than color theory. Most of the colors wheels that followed had 3 primaries (RYB) and 3 secondaries (OGP). After Newton, the most influential color theorist was the poet Goethe with 3 primaries and 3 secondaries.

    Munsell (1898) has 5 primaries and 5 secondaries. His color wheel is closer to the truth in that each primary (red) is located opposite of its true visual secondary (blue-green).

  2. Hi. You mention the “opponent process” theory. I think it’s pretty well established that the signals about color that come out of the retina are encoded as red-vs-green, blue-vs-yellow, and “white-vs-black” or “lightness”. If you ignore the lightness aspect, the combination of R-G and B-Y gives you a two-dimensional space with saturated colors around the outside, and unsaturated in the middle. There are no impossible colors or discontinuities around the outside of that square, so if someone is investigating “colors” they’re implicitly looking at saturated colors, and that’s like walking around a room that has a big pillar or round table in the center. The corners aren’t particularly noticeable, so you’d come out with the impression that colors form a cycle.

    This page http://en.wikipedia.org/wiki/Lab_color_space has some images that give an idea of what that space (including lightness) is like. Purple is definitely there between blue and red. L*a*b* models colors as our brains perceive them in such a way that distances in the space match amount of difference between colors as people report perceiving it. I think it’s Munsell who first systematically built a color model based on that kind of experiment.

    (L*a*b* also suggests why orange seems to deserve a place of its own, whereas RGB goes red, yellow, green..)

    I have to admit the existence of purple always bothered me and I sort of took people’s word for it that the colors indigo and violet existed at the end of the rainbow.

    Physical color is *not* one-dimensional! Light is a mixture of different amounts of *all* the possible wavelengths in the range we can perceive. Each wavelength amounts to a separate dimension, so the physical color space is infinite-dimensional (although a finite amount of energy can only convey a finite amount of information). That infinite-dimensional variety of colors is reduced to four dimensions by the RGBW sensors in the eyes, and then down to three dimensions by the retina’s coding scheme, then down to two dimensions if you’re ignoring lightness, and then a loop if you’re looking for colors as opposed to gray.

    The R, G and B sensors each respond to a different extent to each spectrum line, their responses form three funky overlapping curves http://en.wikipedia.org/wiki/Color_vision#mediaviewer/File:Cone-fundamentals-with-srgb-spectrum.svg . After encoding to R-G and B-Y, the responses to individual spectrum lines arrange themselves in an unevenly-spaced way on a nearly-closed C curve around that perceptual room.

    (The red and green response curves are very similar, but evolution apparently thinks that that difference is very important, and that’s why the yellow space gets stretched enough for orange to have a place.)

    I guess you could say we *confuse* the spectrum with the loop because of that C curve. Maybe that’s why people imagine indigo and violet there. Moving along the C curve, you’re moving in the direction (in the room) towards purple as the blue fades out. The idea that purple is on the rainbow is like the idea that the end of a rainbow touches down somewhere on the earth (which I only clearly realized in adulthood that it doesn’t). The earth and purple are really there, but the end of the rainbow just perceptually points *toward* purple the way it points toward the earth.

  3. As Steve Witham said, the four possible combinations of red OR green and yellow OR blue opponent signals create the 360 degree range of possible hues.

    Violet (i.e. reddish blue) is visible in our spectrum as well as Newton’s, and Newton certainly did not add it to bring the number of colours to seven; because it is in both the five-hue and the ten-hue divisions of the spectrum he described in his Optical Lectures of 1670-72. Dr Burton may be getting confused with the suggestion that has often been made that Newton included indigo in the final seven to fit with his suggested analogy with a musical scale.
    Also, Newton did not give “each of the color as much proportional room on the wheel as they have in a spectrum”, but instead adjusted the intervals to those that the notes of a Dorian modal scale occupy in a circular diagram. See Fig. 7.1.7 and 7.1.8 here:

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: