The Color Of Language: Linguistics Of The Rainbow

The Color Of Language: Linguistics Of The Rainbow

What is color? Is it purely a portion of the electromagnetic spectrum, divisible into nanometers of wavelength and lux of intensity? Or is it a vocabulary that allows us to describe the world around us? Is color art, science, or both?

Is Blue Always Blue?

In 1984, George Orwell invented ‘Newspeak,’ a language that makes alternative thinking impossible by removing the words used to describe such thought: if you have no word for ‘revolution,’ you will not start one... Newspeak was based on the idea of ‘linguistic relativism,’ the Sapir-Whorf hypothesis. Anthropological linguist Edward Sapir and his student, Benjamin Whorf, were convinced that our language constructs our reality: that we see the world through the lens of our own language and anything not encompassed by our language is – to us, at least – unthinkable. Do we live within the confines of our own linguistic reality?

Color terms have long been a favorite testing ground for proponents and opponents of linguistic relativism alike. The color vocabularies of the world’s languages are, well, colorful, and far from identical. Russian discriminates between ‘light blue’ goluboy (голубой) and ‘dark blue’ siniy (синий). Dani, an Indonesian language, has but two words for color: mili, usually associated with dark colors, and mola, usually associated with light colors (it is more complex than this, but that’s the gist). Yet despite these fun linguistic anecdotes, generally speaking, we all share the same color palette. In the late 1970s, the World Color Survey looked at 110 languages from non-industrialized countries worldwide (it is thought that color saturation in industrialized nations skews results for languages like English and French). The survey found that when all the data was plotted, six cross-linguistic peaks emerged, corresponding to English’s pink/red, brown, yellow, green, blue, and purple. Some peaks were taller than others, and some languages had color terms that did not fit into the major peaks, but the survey provided evidence that we’re all more or less looking at the same rainbow.

Photo by -sel-

Why is Blue 'Blue?'

Human eyes have two kinds of photoreceptor cells: rods and cones. Rod cells have one type of photosensitive pigment that allows us to differentiate between light and dark and helps us detect motion. Cone cells have three types of photosensitive pigments – red, green, and blue – that allow us to see in color and in detail. Together, they tell us everything they see in the visible spectrum. But biology is only half the equation. When you look at something – the sky, for instance – your rods and cones set in motion a complex psychological process that enables you to describe what you see. This is true for all stimuli, but we’ll focus on color here.

So let’s look at the sky and see what happens. Step one is perception: your rods and cones take in the color. They tell your brain that they have perceived reflected light with a wavelength of, say, 465 nanometers. Step two is categorization: you must place what you see along the visible spectrum. Your brain says this is BLUE (all caps means it is a color category, not a color itself). Step three is lexicalization: you put that category into words: “The sky is so blue today!” The lexicalization process allows for both synonymy (RED includes both crimson and carmine) and polysemy (teal falls under both the BLUE and the GREEN categories).
But what about the Russians? Or the Dani in Indonesia? We know that neither has a word for the BLUE category, but do they still have the category?

Is My Blue Your Blue? The Theories:

1. Genetic Determinism: My blue is your blue because our biology says so

In 1969, anthropologist Brent Berlin and linguist Paul Kay joined forces to determine if there was any order to the chaotic world of international color. By looking at basic color terms in languages around the globe, they defined what they called an ‘evolutionary order’ for color terminology. All languages, they found, have at least two categories for color: WHITE and BLACK (more like ‘light’ and ‘dark,’ really). After that, each linguistic group adds colors to their language in relation to the needs of their local environment, in this order:

  • White/Black
  • Red
  • Green/Yellow
  • Blue
  • Brown
  • Pink/Purple/Orange/Grey

Berlin and Kay assumed that everybody has the same colors, and as a language becomes more ‘advanced,’ it adds color words in this order. Needless to say, many people were unhappy with this approach that clearly labeled Indonesian tribesmen as ‘backward’ and Western Europeans as ‘advanced.’ It has also been argued that if color terms do indeed ‘evolve,’ then BLUE, the color of both the sky and bodies of water, should be more universally salient. While there is general agreement that all humans see the same color spectrum (with slight variation from person to person, of course), there is still no consensus on whether we all see it the same way.

Photo by mscolly

2. Empiricism: My blue is your blue because we taught ourselves so

In direct contradiction to the determinist idea that humans are innately endowed with color categories, it seems that people can actually acquire new sets if taught. It is thanks to a psychologist named Eleanor Rosch that we know so much about the Dani people of Indonesia and their limited color vocabulary. Rosch wondered whether, with their mili dark colors and mola light colors, the Dani still had other psychological focal points on the color spectrum, despite not naming those categories. She found that, indeed, they did. When taught ‘new’ colors, the Dani were able to learn them, though they were able to do so better for focal colors (those that appear in many of the world’s languages, such as ‘red’) than for non-focal colors (such as ‘maroon’ or ‘fuchsia’). In short, the processing system for color categories appeared to be innate in the Dani, but how members of the tribe represented the categories was the result of individual learning.

3. Linguistic Relativism: My blue isn’t your blue at all (if we speak different languages)

And then, of course, there is Sapir-Whorf and ‘Newspeak.’ According to the tenets of linguistic relativism, one’s language should define his or her categorization of the color spectrum. In other words, since the Dani have no word for green or blue, they shouldn’t be able to discriminate between GREEN and BLUE. It’s a beautiful thought that our language could twist and shape the world around us, giving each linguistic group a truly distinct perspective on the same stimuli and phenomena...but it just isn’t so. We all see with the same eyes. And yet, some linguistic relativism does seem to be at play in our categorization of color. The Dani, though able to discriminate between GREEN and BLUE, did not do so as well as Americans (whose language has two separate categories). Another psychologist, Jules Davidoff, and his associates, replicated the Dani experiment with the Berinmo tribe in Papua New Guinea, who have separate categories for YELLOW-GREEN and BLUE-GREEN. They found that the Berinmo were better at discriminating between yellow-green and blue-green than Americans were. But Americans, who have distinct BLUE and GREEN categories, were better at discriminating between blue and green than the Berinmo were. Language obviously plays a role, but it has to contend with biology as well.

Photo by andrewpaulcarr

So what is blue? Conclusion

One of the youngest linguistic color researchers, Tony Belpaeme, has used computational modeling to test all three theories: genetic determinism, empiricism, and linguistic relativism. After years of research, he has determined that physiological, environmental, and ecological constraints are not enough to lead all individuals to the same color conclusions: we don’t just all happen onto the same focal color categories. Instead, both genetic determinism and linguistic relativism help us break down the spectrum; there does seem to be something psychologically inherent about color, but there is a fairly large area of flexibility within which language can play.

In short, my blue is not BLUE because I taught myself so. My blue is BLUE because my cones and rods see a specific wavelength that is psychologically-salient to me, and my culture has tuned me to it by giving it a name.

So next time you look at the rainbow, remember: to some people, BLUE isn’t blue at all.


Belpaeme, Tony, “Factors influencing the origins of colour categories”

Belpaeme, Tony, “Understanding the origins of colour categories through computation modelling,”

Belpaeme, Tony and Joris Bleys, “Colourful language and colour categories,”

Olin, Dirk, “The Way We Live Now: 11-30-03: Crash Course; Color Cognition,” New York Times, November 30, 2003 (accessed May 12, 2008).

Regier, Terry and Paul Kay, “Color naming and sunlight: Commentary on Lindsey and Brown (2002)”

Title photo by christyscherrer: “Evolutionary Rainbow” mural by Yana Zegri in San Francisco.

jessica_icon.jpgAbout the Guest Author, Jessica Alexander
Jessica Alexander is a writer, translator, and hopeless devotee of overstuffed dictionaries. For more titillating linguistics, check out dailycharacter. Or, if you just want to send her love letters…

Showing 1 - 14 of 14 Comments
Great article. I felt I was in high school again as this reminded me of Merleau-Ponty's philosophical theories on this subject. Nice. :)
Anthropological linguist Edward Sapir and his student, Benjamin Whorf, were convinced that our language constructs our reality: that we see the world through the lens of our own language and anything not encompassed by our language is – to us, at least – unthinkable. Do we live within the confines of our own linguistic reality?

oooohh... yesss....

This webpage we're looking at is constructed of language. That language is interpreted by the server and your browser to create what you see on the screen. We've been creating the new universe of the internet for several years now by writing.

Going further on this, I'm not a strong programmer, but I get the gist of Object Oriented Programming and how it can be used to create whole environments with parameters defining gravity, walls, atmosphere, character constraints, hit-tests, etc. Perhaps our entire universe is just one big program written by an omnipotent geek. *que Keanu Reeves' "woaahhh"*

And God said, "Let there be light"; and there was light. - It had to be spoken to become.

Great article, quite stimulating. I'd to see more like this.
Diving into another rabbit hole:
this WAS cool. never thought about it. i just thought people were like, "oh, that's blue," instead of BLUE>dark blue>navy>etc. and i knew some people didn't have names for colours - but does it matter? i mean, in whatever environment their in, couldn't they not have, say, fuschia? or something like that? couldn't it just be RED? or PINK? but whatever, this is really interesting.

oh, a note to serostar:
"I know kung fu..."
I'm so glad you all enjoyed the article!
I remember being told once that the human eye can only see 16 shades of grey and thinking "that's total BS." I think it'd be interesting to test COLOURlovers vs other computer users on their ability to distinguish between colors. You'd have to cheat-proof the test somehow, though. No eyedroppers allowed.
Great article, thank you!
Very good article. One of my Czech friends (the one teaching me the language) and I have ongoing discussions over colour definitions: my "azurova" (cyan) differs greatly from hers. "Purpurova" and "fialova" (both listed as: purple, violet, or magenta, depending upon whom you consult) mean very different things to her. It's interesting to me, but then, I can't necessarily agree with native English speakers on what "purple" means, either. :/
I always just liked blue, A very good article
well. back in school... i always managed to fall asleep studying color theory. but this article just kept me intrigued! actually, everything on this site keeps me reading everyday!
Thanks, all!

Post a Comment

You must be logged in to post a comment.