20171126

Is violet found in the rainbow?

We are taught that Richard Of York Gave Battle In Vain and for this gem we have to thank Isaac Newton who also had a whacky theory of how colour related to music. Whilst we can credit Isaac with some understanding of colour theory, his seven named spectral colours have been grossly misused over the centuries since.

Isaac's colour wheel

And so Johnny is taught in school that the rainbow has seven discrete colours. He is also falsely taught that the primary colours are red, blue and yellow. As a child I quickly learned from experiment the falsehood here. You try getting a decent purple by mixing red and blue paints! It would be more correct to say that the primary colours for additive mixing are red, blue and green (but c.f.) and that the colours you get from mixing these in equal proportions, aka secondary colours, namely yellow, cyan and magenta, are for this very reason the primary colours for subtractive (paint) mixing.

In our school room here we have a colourful set of of containers which I have pictured below, Which of the container colours shown are not found in the spectrum of white light?




A friend JHA recently gave an dissertation on blue and red, as in the colours of the rainbow. He excluded purple and did not mention violet thus implying that the ends of the visible spectrum are red and blue. I consider this a gross injustice, I feel as though I have been robbed. Whilst I can do away with indigo because the truth is that the spectrum is a continuous variation in colour, to say the rainbow ends in blue is a blatant lie. But the truth of the matter is not so clear. I thought JHA would go on to talk about green as the third "primary colour" and thus disprove the law of the excluded middle but the point he was making was somewhat different. But whatever he was talking about, he very definitely did not mention spectral violet.

So what is the truth? All sources agree that human eyes are trichromatic and thus possess three types of colour sensor or "cones" which have peak spectral responses around blue, green and red, and this is why colour TV and film can reproduce most colours using pigments of these colours aka primary colours. Although some folk, 10% of men I am told, only have two types of cone and are thus "colour blind". There is even the suggestion that some women might even have a forth type of cone which I regard as wholly unfair.  So far so good.  But hereafter theories differ.

There are those who deny that there is a problem at all. Until I hear him recant I must place my friend JHA in this category. For such people violet cannot exist in the rainbow.

Some say that the the "blue" cone is a misnomer because it actually has peak response to what we call violet spectral light. Spectral light of slightly longer wavelength actually tickles both the violet cones and to a lesser degree the other cones, and this combination we perceive as blue. And thus the sky is actually violet.

Others (and here) say that in fact the "red" cones have a secondary response in the violet area and thus light of this wavelength tickles both blue and to a lesser degree red cones and we perceive this mixture as violet.

Both parties agree that the eye perceives other mixtures of red and blue as variations of the purple / violet / mauve / magenta colours some of which are rather poorly defined.

In the second and third theories violet is definitely a spectral colour as distinct from blue, so that the only reason scientists might talk about the blue and red ends of the visible spectrum is because blue is an easier concept than violet. After all the first meaning of the word is a small wild flower.

One more (at least!) point that needs to be made is that, even in the rainbow, we are typically not looking at pure spectral colours, by which I imply monochromatic light. You'll get a better sensation of monochromatic light aka pure saturated colour by looking down a spectrometer, a device that splits light into a rainbow of colours e.g. using a prism but uses lenses and a slit to ensure that the eye is looking as only a very narrow angle of the emitted spectrum. As most of us do not have access to a spectrometer, try looking directly at a prism splitting white light. A chandelier crystal will do.

Colour film and TV can never do full justice to the eye's ability to see colour. Indeed the eye itself can never do full justice to the colours actually implicit in the spectrum. In either case the issue is firstly that the eye is trichromatic and secondly that the red, green and blue cones each respond to a broad range of colours and thus, apart possibly from the extreme ends of the visible spectrum, any particular spectral colour is bound to stimulate more than one type of cone and thus results in reduced saturation. This effect is progressively more evident as you move away from the centre of an oil film spectrum as captured in my recent photo.  Here you will see colours verging on brown, a colour no-one to my knowledge has ever claimed was in the rainbow.



2 comments:

  1. I am well aware of the ambiguity surrounding the nomenclature; in modern scientific terms, magenta is the color found on the wheel but not on the spectrum.

    I was speaking to the lay person (not designers or color theorists) who does not understand that terms like magenta and cyan have precise color values. "Purple" had to suffice for that audience. (Most have no idea that cyan is even a color!)

    And I know better than to claim violet isn't on the spectrum.

    For a further dive into how nomenclature affects perception, take a gander at how the ancient Greeks "had no word for blue" http://www.businessinsider.com/what-is-blue-and-how-do-we-see-color-2015-2

    For a fun optical illusion that tricks the brain into seeing purer cyan than "exists", see http://themetapicture.com/youve-never-seen-true-cyan-before-and-its-beautiful/

    ReplyDelete
  2. Thanks for your input and interesting links: I now rank you suitably recanted!

    ReplyDelete