Epilogue

---

I left copies of the preceding article in the physics department lounge at Stanford University. One reader was kind enough to send me a response.

He told me that the idea of red and green combining to make yellow had never made any sense to him, either--until he researched the phenomenon and found that it was actually based on a characteristic of our color vision, rather than a property of light itself.

In actuality, he explained to me, yellow light has its own distinctive wavelength and frequency, as does every other color of light; and therefore no two pure colors of light can ever be said to truly "make" any other color.

But our eyes' color-sensing mechanisms, the cones of our retinas, react to only three types of light (red, blue, and green), and "infer" other colors based on the relative degrees to which the cones for red, blue, and green are stimulated. Thus, if our receptors for red and green are excited simultaneously, we perceive yellow. This is really the only sense in which red and green can be said to "produce" yellow, he wrote.

Being a rather cussedly stubborn sort, I wrote back that I still wasn't entirely satisfied that the conventional RBG triad had been vindicated by his explanation.

First of all, I didn't like the idea that the perception of yellow that's induced when red and green light are combined depends on a quirk of our physiology. What if I created an artwork out of colored light in which I used red and green to induce the viewer to perceive yellow, but the viewing were actually done by someone on a spaceship from Mars, with a different means of sensing color?

(While I considered such a visitation unlikely, and the possibility that an extraterrestrial visitor might want to see my artwork even more unlikely, I still couldn't help being troubled by the somewhat gimmicky, physiology-dependent, and anthropocentric nature of the effect. It struck the same kind of discordant note in me as that line of reasoning by which some people argue that if a tree falls in the woods where no one can hear it, there is no sound.)

I also questioned whether it might actually be preferable, when dealing with colored light, to use a set of primaries that were deliberately impure, with each representing a third of the spectrum. My hunch was that perhaps when related colors of light from the same third of the spectrum are shown together, they might somehow "average out" to the color in the middle of the band.

I had the idea that maybe if two colored lights of this type were combined--especially in cases like that of firetruck red and green--that there would be enough actual yellow light included in the mix, and that the yellow would be close enough to the "spectral average" of the admixture, that it could be said to truly be the predominant hue of combined light group, and not just a physiological trick based on the foibles of retinal cones. Since no two colors of light truly made a third color, I felt it might be preferable to have the resulting admixture at least contain the color that was said to be produced.

I didn't receive a second reply from this correspondent. More than a year passed before I gave the matter any significant consideration again.

Then one day out of the blue, I had a striking sense that I had asked the wrong question. What I should have asked, I now believed, was simply this:

If no wavelength and frequency of light can produce any other wavelength and frequency, what point can there be in designating any of these wavelengths and frequencies of colored light as "primaries?"

I can see very good reasons for designating some pigments as primaries. If you mix cyan, yellow, and magenta, you truly can produce all the other colors, and without having to exploit goofy quirks of the human eyeball. To me, that makes them good and useful building blocks that are simple and based in objective reality, rather than just the tissues by which we sense reality.

But where is the sense in designating any color of light a "primary," if no admixture of colored lights can truly make another color of light? To me, this deficiency renders the whole concept of primary colors meaningless, except when we're talking about pigments.

To me, this realization seems like it ought to set the record straight once and for all about what's really a primary, and what isn't. Blue and yellow and (printer’s) red are the primary colors—and that’s that.

Pretty much just like Mrs. Finkey said.

But of course, I'm not a physicist, and therefore can't really expect anybody to take what I say seriously...

© COPYRIGHT 1993 ROBERT WINTER.  ALL RIGHTS RESERVED.