A Proposal for a Better Rainbow

I learned the rainbow from Mickey Mouse Clubhouse.

The song went, "Red, orange, yellow, green, blue! Don't forget there's purple too! Rainbow colors for me and you!" This was the rainbow I was used to, so when I heard the classic "Roy G. Biv" rainbow on PBS Kids, I was not a fan. The color 'indigo' did not seem like a useful one at all, and felt like an unnecessary addition. First of all, it brings the number of simple colors up to 7, an odd number, but also, even as a kid I knew there was very little space already between blue and purple on a spectrum, and packing in indigo, which really just looked dark blue, felt unnecessary.

However, the origins of indigo on a rainbow are more interesting than I was expecting. The ROYGBIV rainbow was coined by Isaac Newton, in his Opticks, and I used to imagine that his distribution mapped onto the visible spectrum of light looked something like the messy diagram pictured below (note that this is not a color wheel, but just a spectrum of visible light, as Newton would have seen with his glass prism.)

Newton's rainbow according to our modern conception of his color names.

As you can see, the colors do not seem very well distributed. I flipped through Opticks trying to understand exactly why he added indigo to his list, and I think there are two main reasons. First, the number seven was regarded as a perfect, or "holy" number, and this clearly had an influence. On page 144, he writes, "[I] distinguish its circumference into seven parts...proportional to the seven Musical Tones." But why indigo? If you must divide the colors into seven, why not a color between green and blue, like cyan? I can find two potential reasons for this. First, cyan was not an easy color to find naturally, nor one most people in Newton's time would have been familiar with. It is easy to simulate with computers, or modern synthetic dyes, but is rarely naturally occurring. The second potential reason came from this excerpt from Wikipedia:

"According to Gary Waldman, "A careful reading of Newton's work indicates that the color he called indigo, we would normally call blue; his blue is then what we would name blue-green or cyan." If this is true, Newton's seven spectral colors would have been:"

What Newton may actually have been picturing

Even this interpretation is controversial, due to the human eye having a hard time distinguishing between Newton's "indigo" and "violet" when looking at a real rainbow.

If these are the colors Newton was envisioning when he created the ROYGBIV rainbow taught today, then we have continued to use a rainbow that is no longer accurate to our modern conceptions of color names. There was no color printing in 1704, so we will never know for sure, but since our modern conceptions of blue and violet are already so close together on the visible spectrum, that Newton would try to squish indigo in between them seems unlikely, and thus I agree with Gary Waldman on this point.

So now we come to the present; or, at least, to the early 19th century, when the first artists' color wheel was developed. In schools and art classes, we are used to seeing color wheels that look like the image pictured right. While missing indigo, it uses ROYGBP, the order I learned from Mickey Mouse. But there is an important discrepancy here: this commonly-taught rainbow is not accurately weighted to real-life colors.

A classic color wheel

The unweighted color spectrum

Pictured left is a real rainbow color wheel, accurate to all the colors our brains are capable of perceiving, spaced evenly by wavelength (plus magenta). Ignore the difference in color saturation and just look at how the hues are laid out. Starting at 11:00, to around 3:00, are what are generally considered to be the warm colors, and everything from 3:01 to 10:59 are cool tones, whereas on the traditional color wheel above, the warm/cool divide is a straight line through red-purple and green-yellow. See the difference pictured below:

Both spectrums have been overlaid with yellow and blue to show the approximate warm and cool sections. Left is the actual color spectrum, and right is the traditionally taught color wheel.

So this brought to my attention something I hadn't thought about before: the unbiased, unweighted color spectrum is not pleasing to the human eyes. We seem to prefer more warm colors than occur naturally in the world, and while there are proposed explanations for this, I'm not going to go into it here.

Subtractive primaries

I grant that the classic color wheel is aesthetically pleasing to the eye, but it's worth noticing that it does not feature cyan or magenta. Scientifically speaking, magenta is not a "real" color, as it does not have its own wavelength, but is rather a sensation produced by our brains when our eyes receive blue (or violet) and red light simultaneously. Cyan is likely missing simply because, as noted, it is a rare color to find in nature, and is too harsh a color for most paintings anyway. The same could be said of magenta.

As a kid, however, I was taught that the primary colors were yellow, red, and blue. This is objectively false, at least by our modern conceptions of what those color names mean, though maybe not by Newton's. The subtractive primaries, as pictured to the left, are cyan, magenta, and yellow, or CMY. These are the pigments that printers use to produce any color, and they are the visually brightest colors on the spectrum.

When pictured in black and white, so that you can see only the colors' brightnesses, the subtractive primaries are all brighter than the colors they mix to create. This makes sense in the context of printers or painting; you cannot mix together two colors to make a brighter color. Note that printers also include black to save colored ink.

Subtractive primaries in B&W.

The additive primaries, however, are red, green, and blue, or RGB. These are the colors your digital screens use to display color, and can be seen pictured left. Now, the reason screens do not use CMY is for the same reason printers don't use RGB. Screens emit light, while pigment reflects or filters it. If I print cyan dots on a page, you will see cyan. This is because the white light (an even mixture of many colors) hitting the paper will hit the pigment dots, and all the colors in that white light except cyan will be absorbed, while the cyan reflects back into your eyes. With screens, on the other hand, the light itself is being created and emitted. Magenta and cyan dots on paper will make dark blue, because more light is being absorbed, and less reflected. But red and blue light being emitted from a screen will create magenta, because more light is hitting your eyes. Additive primaries add light to create brighter colors, subtractive primaries subtract light to create darker colors.

Note the clear bands of brightness

And so, I propose a new rainbow to teach in school: red, yellow, green, cyan, blue, and magenta (RYGCBM), as pictured right. Not only does this new rainbow accurately represent the visible color spectrum, but it sorts the colors into the additive and subtractive primaries, alternating, so that all the colors named are evenly spaced.

Orange and purple are gone, but they are, in my opinion, "in-between" colors, and do not deserve distinction when talking about the basic rainbow. Pictured left is the RYGCBM rainbow, with a black and white filter, so you can see the brightnesses of the colors. Each red line highlights a dark or light streak on the wheel. Orange and purple have no such distinction, which is why I call them "in-between" colors.

Is my RYGCBM rainbow really worth adapting?

After all, I grant that it removes the aesthetic preference shared by humans of splitting the colors evenly into warm and cool, even if that isn't accurate to the natural world. Further, it leaves confusion as to where orange went, in a world where orange (and brown) are commonly distinguished colors. To be clear, I'm not suggesting we eradicate all colors except for these, simply that, for the purposes of art classes and color theory, the RYGCBM rainbow is the most intuitive to teach. It should not be used to find complimentary colors, or pick palettes for a painting, because traditional color wheels are designed specifically for that purpose.

However, the Roy G. Biv rainbow taught today is an outdated metric, not only because it has been surpassed by our modern ability to display and understand color, but because we likely do not even share the same conceptions as Newton of what those color names actually represent.

Page updated

Google Sites

Report abuse