Friday, October 23, 2009

The high Berlin-Kay rank of yellow

In a previous post I wondered why, given that our color vision uses an RGB system, our languages consistently treat the non-primary yellow as more basic than blue and sometimes more basic than green. Well, sometimes you have to step back and question the givens.



The above diagram shows the sensitivity of the three cone types to different wavelengths of light. (It comes from Wikipedia. All I've added is the vertical lines dividing the spectrum into "colors.") Once you see that the sensitivity of the so-called red cone actually peaks at yellow, the latter color's status as an honorary primary makes more sense.

Now the question is, why is red considered so much more basic than violet? Being both a bookend of the visible spectrum and the region where the "blue" cone's sensitivity peaks, violet ought to be as important as red and yellow put together -- but, while red is universally treated as the very most basic spectral color, violet is not a basic color in any language. (Purple sometimes is, but purple is not the same as spectral violet.) In fact, violet is so unimportant in our color vision that it's the one spectral color your computer monitor is physically incapable of displaying -- and no one notices.

3 comments:

  1. This morning I was just thinking about this sort of graph, and was going to post a comment saying you should look at it -- but you beat me to it.
    There are a lot of strange things about that graph. I wonder why we precieve such a wide section of wave lengths are green, while such a narrow band as yellow? And you are right, violet is just strange...

    ReplyDelete
  2. One thing this graph shed light on for me is the blue-gray connection. Languages without a word for gray almost invariably refer to that color as "blue" -- but, pure gray is a completely neutral color, composed of every wavelength in equal parts. Why should we consider it closer to blue than, say, to red? But looking at the above graph, you can see that the "blue" area is where the three lines come closest to converging -- where all three types of cone are being stimulated to roughly the same degree.

    Even more interestingly, this gray area is near the border of blue and green -- the cyan region. Perceptually, cyan (especially low-intensity cyan) is indeed so close to gray that some people -- including Vivian and our father -- can't tell the difference.

    I think I'm going to replot the graph, measuring the stimulation of each cone not in absolute terms but as a percentage of the total, and see if that clears anything up.

    ReplyDelete
  3. What do you mean when you suggest that computer monitors can not display a viable visible Violet...???
    Could you give an example of something that exists in nature, that is violet, And can be viewed as violet, that a computer image can not accurately reproduce...! ( ??? )

    ReplyDelete