The news website Vox has produced some good videos on linguistic topics, which can be found amidst their many other clips. Its latest one looks at the vexed question of colour names and categories in different languages, and in 6½ minutes it offers a decent summary:
*
*
The video touches on Homer’s wine-dark sea, anthropological racism, and recent research revising the colour sequence made famous by Berlin and Kay’s influential work. It doesn’t mention Lazarus Geiger, the German philologist who proposed essentially the same sequence a century earlier – but then, it is just an overview.
For more on the subject (and other aspects of linguistic relativity), Guy Deutscher’s book Through the Language Glass is a balanced and enlightening read.
Sentence first 
My masters honours research project and dissertation was on early British encounters with the native languages of Australia, especially in the Sydney area. I noted that that five colours recorded by Marine Lieutenant Williams Dawes were black, white, green, yellow and red. This is not to say that they didn’t have more names, rather just to say that those were the ones he recorded. I had Berlin and Kay at the back of my mind, but didn’t specifically cite them.
The video states that blue was uncommon until the industrial era. But I can see a blue sky out my window, and Sydney is on a harbour fronting an ocean, so the people here must have been surrounded by blue (and green) all day every day.
Blue is a peculiar case, so strangely absent from Homer that some researchers suggested that a physiological or cognitive defect may have obtained at the time. Gladstone, Geiger, and Rivers all wondered about this. In Deutscher’s book Through the Language Glass he describes an innocuous experiment he carries out with his daughter, Alma: he doesn’t mention the colour of the sky to her, in order to see how quickly she can identify it for herself.
He says she recognised blue objects from the age of 18 months, and began using the word (‘boo’) a month later. Asked to identify the colour of prototypically, artificially blue objects, she had no difficulty, but when asked the colour of the sky (when it was particularly blue), ‘she would just stare upwards in bafflement’. At 23 months she finally answered the question – and decided it was white. A little later she said blue, then white again, etc. At four years old, she said a pitch-black night sky was blue. So the colour of the sky (or indeed the sea) is not as obvious or straightforward as our culturally assured adult minds find it. Deutscher concludes:
Interesting. I’ve seen that book at bookshops but my ‘to buy’ list is longer than my budget, even if language-related books are tax-deductible.
I’ve read that (some) N. American Aboriginals called the European settlers “white eyes”. That might have reflected the blue-white issue.
On the northern plains a Native name for the settlers was “Wasichu”, which I think means “those who are so many”.
“Wasichu” was colorless and could apply to anything in large numbers, such as the buffalo.
Regarding racism and blackness: when the 2nd Iraq War was gearing up, Saddam Hussein, indifferent to or unaware of any incorrectness, remarked that the White House ought to be called the Black House — presumably because its intentions toward him must align it with the Prince of Darkness. That is, that color bias is probably universal.
Interesting note on the blue–white question.
Black is certainly associated with moral darkness in many cultures, even if only by negation of the pervasive God=light symbolism, where white represents spiritual illumination. But the association isn’t universal: in many contexts, black is more neutrally associated with the absolute (like white). In some Eastern religions and mythologies black is the colour of water, or certain divinities, such as Kali and Mahakala. And even in Christianity, black can have positive meaning, symbolising the renunciation of worldly goods and concerns: hence the colour of priests’ garments.
Interesting, but the bit at the end about presenting multiple colours to artificial agents didn’t make sense to me. Something crucial is missing.
See Language Log: http://languagelog.ldc.upenn.edu/nll/?p=32799 (a recurrent neural network names colours)
and kxcd https://blog.xkcd.com/2010/05/03/color-survey-results/ (kxcd readers name colours).
[…] fascinating short video explains how different languages have treated the process of categorising […]