4 ms·

How we gave colors names

105 points by sid6376 14y ago

hayksaakian 14y ago
The xkcd article linked within is more interesting than the actual article.
- tspiteri 14y ago
  I found both really interesting. Whereas the xkcd article (http://blog.xkcd.com/2010/05/03/color-survey-results/ http://blog.xkcd.com/2010/05/03/color-survey-results/) goes into some detail into one experiment, the original article has a wider scope and touches on several experiments, the xkcd experiment being one of them. Also interesting is the follow-up article about how language is used by the brain to distinguish between different colours (http://www.empiricalzeal.com/2012/06/11/the-crayola-fication-of-the-world-how-we-gave-colors-names-and-it-messed-with-our-brains-part-ii/ http://www.empiricalzeal.com/2012/06/11/the-crayola-fication...).
alex-g 14y ago
This blog article (http://korystamper.wordpress.com/2012/08/07/seeing-cerise-defining-colors/ http://korystamper.wordpress.com/2012/08/07/seeing-cerise-de...) by lexicographer Kory Stamper raises some similar issues about color names in English - how does one define "cerise" or "taupe" in the dictionary, when nobody can agree on what precisely they mean (where to draw the boundary). There is also the issue of how best to write the definition so that people can actually understand it. Webster's Second and Third New International Dictionaries included color plates (which seems reasonable!) but also gave HSB values in the Second and complex color comparisons in the Third (cerise is "a moderate red slightly darker than claret, slightly lighter than Harvard crimson, very slightly bluer and duller than average strawberry, and bluer and slightly lighter than Turkish red"). For the Learner's Dictionary, though, they used definitions like "green is the color of grass", less precise but probably still having some psychological truth - the color of grass is surely an anchor point for our concept of green.
manveru 14y ago
I was really mystified why Japanese kept calling a green traffic light "blue" (in Japanese, that is). It didn't really make sense to me until reading this, so thanks a lot.
- yen223 14y ago
  I was mystified by 'redheads', since I've never seen anyone with naturally red hair. The article also did help clear that up too.
JangoSteve 14y ago
This post was really interesting and made me think a bit differently about color names (I think to an even deeper extend than the xkcd survey post from a couple years ago). I also found the first comment [1] quite interesting, namely this part: For example, ao in Japanese comes from the dye plant, ai, which as a dyestuff covers the whole of the blue-green portion of the spectrum. Some cultures have what might at first seem to be peculiarly chosen "basic" color names until you learn their associations with the culture's central food sources or dye plants, or precious commodities. The color name that covers both blue and green in many native languages of the American Southwest is also the name for the stone, turquoise. (In cultures where a staple food is poisonous when green and edible when red, you can be sure there are names for green and red.) In our own history, we have a very similar example to the "blue" traffic lights of Japan: "orange" didn't enter English as a color name until the 16th century, after the fruit itself was first brought to England, quite late in the evolution of our color vocabulary, which is why we still refer to "red" hair. [1] http://www.empiricalzeal.com/2012/06/05/the-crayola-fication-of-the-world-how-we-gave-colors-names-and-it-messed-with-our-brains-part-i/#comment-567420008 http://www.empiricalzeal.com/2012/06/05/the-crayola-fication...
- jballanc 14y ago
  What's funny about that comment (and a handful of others that follow) is that they completely miss the fact that "turquoise" is probably one of the most prominent examples of a "made up" color name coming from the French for "Turkish".
Fargren 14y ago
On a similar vein to that of the japanese traffic lights, here in Buenos Aires traffic lights for people used to be green and red, but at some point they changed to white and red. However, people still say "verde" (green) for the white light that means you can walk. Traffic lights for cars are still green, yellow and red.
tokenadult 14y ago
I've been running into this issue since the 1970s as a second-language speaker of Chinese. My wife, a native speaker of Taiwanese (whose parents, like most Taiwanese persons in their generation, were educated in Japanese under the prewar occupation of Taiwan by the Japanese Empire), tends to this day to confuse "green" and "blue" when speaking English. Those colors are both 青 in her mind. [AFTER EDIT, reply to first kind reply below: I mention Japanese that my wife's parents spoke because it was the language focused on in the submitted article, and to note that in their generation no other second language that might have introduced a distinction between green and blue was spoken in the family home. You are correct that historically Chinese also used 青 as a color term with a semantic range including both green and blue from the point of view of an English speaker.] But another puzzler I encountered as I learned Chinese, first in the United States and then in Taiwan, was the broad range of colors that would be identified as 黃 (traditionally translated "yellow," a term definitely used for the yolk of an egg). One day in Chinese class in Taipei, my teacher, an older (birth decade 1930s) native speaker of Mandarin who grew up in Beijing, referred to the wooden tabletop in our classroom as 黃 in color. Aha! The term 黃 covers the full range of not just yellow, as we refer to yellow in English, but also pretty much the entire range of what English speakers call brown. Of course. Now I understand why the "Yellow River" 黃河 is called that, even though when I have seen it directly the silt in the river made it look brown to my eyes, not yellow. And similarly for referring to grass browning ( 變黃 ) in autumn, as to my English ear, I would not call the color of grass in autumn "yellow" but rather brown. The book Basic Color Terms: Their Universality and Evolution by Brent Berlin and Paul Kay http://www.amazon.com/Basic-Color-Terms-Universality-Evolution/dp/1575861623/ http://www.amazon.com/Basic-Color-Terms-Universality-Evoluti... was originally published in 1969, and it set the agenda for subsequent studies of how color terms vary across languages. The conclusion (as also related in the blog post kindly submitted here) is that most languages around the world begin with a very basic set of color terms that gradually accumulates more fine distinctions, in a generally invariant order across most world cultures. English has many color terms, but not quite as many BASIC color terms as a few other European languages. We can see this same kind of process in older terms and phrases in English. What is called "red hair" in English (my late dad had that hair color) would more likely be called "orange hair" today, except that there was no word "orange" in English (that comes from the name of a foreign fruit, after all) at that time, so the word "red" had the full semantic range of today's word "red" and today's word "orange." That's quite comparable to the situation in modern Chinese, where there now are competing, very new, words for "brown" ("[coconut] palm-colored" or "coffee-colored" often being used in actual daily conversation) but many historical phrases in which 黃 ("yellow") is used for colors in the same range.
adorable 14y ago
Fascinating article. For a long time I've wondered why we humans usually describe colors as a 3 dimensional space (RGB) where it really only is a 1-dimensional one (wavelength). Took me sometime to realize this was because we have 3 types of rod cells ... and to start wondering how the world looks for the living creatures who have 4, 5 or more rod cells...
- bonobo 14y ago
  Ask a tetrachromatic woman: http://en.wikipedia.org/wiki/Tetrachromacy#Possibility_of_human_tetrachromats http://en.wikipedia.org/wiki/Tetrachromacy#Possibility_of_hu...
- Someone 14y ago
  We humans? More 'we computer programmers'. The model 'humans' have more closely follows CIE (http://en.wikipedia.org/wiki/Color_model#CIE_XYZ_color_space http://en.wikipedia.org/wiki/Color_model#CIE_XYZ_color_space). If, instead of measuring it, you asked naive subjects (probing for what they are conscious of), you might even get answers where red and violet are highly separated (as in a rainbow), but I think you can make them question that model easily by creating a palette going from red to violet that does not include orange, yellow, green, or blue. In either case, the problem remains that humans perceive colors that do not physically exist, such as brown (aka 'dark orange-ish') and gray (aka 'shades of white') Historically, some people would have known about mixing paints. That might have change their ideas about color, but that isn't RGB, either. Finally, I am not sure one's color space will change with the addition of types of cone cells (it is cones, not rods that are color-sensitive. It alliterates: kones are for kolor).
- andrewcooke 14y ago
  i understand what you meant, but, in a sense, if we had "perfect" colour detectors then (ignoring quantum effects) vision would be an "infinite dimensional space" rather than a 1-dimensional one. unless you are restricted to a world in which the light at any one point is monochrome. maybe we would naturally understand fourier analysis in such a world? it's kind of hard to even think about. maybe monochrome light would appear particularly wonderful (or pure) with such eyes? i was thinking about this the other day when i happened to be sitting in the path of some refracted light - it does appear particularly pure even though, with only three detectors, there's no real reason for it to do so (as far as i can see).
bonobo 14y ago
Now our vocabulary grew past the basic white-red-yellow-green-blue-black separation and I believe it will keep growing. I say this because I consider cyan and magenta to be new ramifications of that spectrum, both colors are well recognized by name today. I think this is due to how information is accessible today. Languages seem to be able to separate a hue in two or more once the actual names are already established—when the burden of naming and categorizing hues gives place to the lesser burden of only memorizing the already named hues, then we can afford to create another new name. Although brands of nail polish and ink have dozens of different names for their colors, they often don't reach a consensus, and the same color can have different names between brands. Without a consensus (language is primarily based on consensus) all these subtly different hues can be said to remain unnamed, I guess... I also wonder if there's a limit to this, to how many colors we can name before considering naming all colors in hexadecimal on a daily basis.
crazygringo 14y ago
Funny, in Brazilian Portuguese nobody refers to traffic lights by their colors at all, despite being the same traffic lights used everywhere else in the world -- the light is either "open" (for green) or "closed" (for red). It took me forever to remember to speak in terms of open/closed instead of in terms of colors, when driving...
- arturhoo 14y ago
  Not quite true. As a Brazilian myself, from the state of Minas Gerais, we most of the times say: the traffic light is green. What is often debated between people from different states is if the light (usually) positioned in the middle is yellow or orange. But yes, sometimes we also say, the traffic light is open/closed.
js2 14y ago
Tangentially related, this reminded me of the color hue test - http://www.xrite.com/custom_page.aspx?pageid=77&lang=pt http://www.xrite.com/custom_page.aspx?pageid=77&lang=pt

105 points by sid6376 14y ago

hayksaakian 14y ago

The xkcd article linked within is more interesting than the actual article.

tspiteri 14y ago
I found both really interesting. Whereas the xkcd article (http://blog.xkcd.com/2010/05/03/color-survey-results/ http://blog.xkcd.com/2010/05/03/color-survey-results/) goes into some detail into one experiment, the original article has a wider scope and touches on several experiments, the xkcd experiment being one of them. Also interesting is the follow-up article about how language is used by the brain to distinguish between different colours (http://www.empiricalzeal.com/2012/06/11/the-crayola-fication-of-the-world-how-we-gave-colors-names-and-it-messed-with-our-brains-part-ii/ http://www.empiricalzeal.com/2012/06/11/the-crayola-fication...).

alex-g 14y ago

This blog article (http://korystamper.wordpress.com/2012/08/07/seeing-cerise-defining-colors/ http://korystamper.wordpress.com/2012/08/07/seeing-cerise-de...) by lexicographer Kory Stamper raises some similar issues about color names in English - how does one define "cerise" or "taupe" in the dictionary, when nobody can agree on what precisely they mean (where to draw the boundary). There is also the issue of how best to write the definition so that people can actually understand it. Webster's Second and Third New International Dictionaries included color plates (which seems reasonable!) but also gave HSB values in the Second and complex color comparisons in the Third (cerise is "a moderate red slightly darker than claret, slightly lighter than Harvard crimson, very slightly bluer and duller than average strawberry, and bluer and slightly lighter than Turkish red"). For the Learner's Dictionary, though, they used definitions like "green is the color of grass", less precise but probably still having some psychological truth - the color of grass is surely an anchor point for our concept of green.

manveru 14y ago

I was really mystified why Japanese kept calling a green traffic light "blue" (in Japanese, that is). It didn't really make sense to me until reading this, so thanks a lot.

yen223 14y ago
I was mystified by 'redheads', since I've never seen anyone with naturally red hair. The article also did help clear that up too.

JangoSteve 14y ago

This post was really interesting and made me think a bit differently about color names (I think to an even deeper extend than the xkcd survey post from a couple years ago). I also found the first comment [1] quite interesting, namely this part: For example, ao in Japanese comes from the dye plant, ai, which as a dyestuff covers the whole of the blue-green portion of the spectrum. Some cultures have what might at first seem to be peculiarly chosen "basic" color names until you learn their associations with the culture's central food sources or dye plants, or precious commodities. The color name that covers both blue and green in many native languages of the American Southwest is also the name for the stone, turquoise. (In cultures where a staple food is poisonous when green and edible when red, you can be sure there are names for green and red.) In our own history, we have a very similar example to the "blue" traffic lights of Japan: "orange" didn't enter English as a color name until the 16th century, after the fruit itself was first brought to England, quite late in the evolution of our color vocabulary, which is why we still refer to "red" hair. [1] http://www.empiricalzeal.com/2012/06/05/the-crayola-fication-of-the-world-how-we-gave-colors-names-and-it-messed-with-our-brains-part-i/#comment-567420008 http://www.empiricalzeal.com/2012/06/05/the-crayola-fication...

jballanc 14y ago
What's funny about that comment (and a handful of others that follow) is that they completely miss the fact that "turquoise" is probably one of the most prominent examples of a "made up" color name coming from the French for "Turkish".

Fargren 14y ago

On a similar vein to that of the japanese traffic lights, here in Buenos Aires traffic lights for people used to be green and red, but at some point they changed to white and red. However, people still say "verde" (green) for the white light that means you can walk. Traffic lights for cars are still green, yellow and red.

tokenadult 14y ago

I've been running into this issue since the 1970s as a second-language speaker of Chinese. My wife, a native speaker of Taiwanese (whose parents, like most Taiwanese persons in their generation, were educated in Japanese under the prewar occupation of Taiwan by the Japanese Empire), tends to this day to confuse "green" and "blue" when speaking English. Those colors are both 青 in her mind. [AFTER EDIT, reply to first kind reply below: I mention Japanese that my wife's parents spoke because it was the language focused on in the submitted article, and to note that in their generation no other second language that might have introduced a distinction between green and blue was spoken in the family home. You are correct that historically Chinese also used 青 as a color term with a semantic range including both green and blue from the point of view of an English speaker.] But another puzzler I encountered as I learned Chinese, first in the United States and then in Taiwan, was the broad range of colors that would be identified as 黃 (traditionally translated "yellow," a term definitely used for the yolk of an egg). One day in Chinese class in Taipei, my teacher, an older (birth decade 1930s) native speaker of Mandarin who grew up in Beijing, referred to the wooden tabletop in our classroom as 黃 in color. Aha! The term 黃 covers the full range of not just yellow, as we refer to yellow in English, but also pretty much the entire range of what English speakers call brown. Of course. Now I understand why the "Yellow River" 黃河 is called that, even though when I have seen it directly the silt in the river made it look brown to my eyes, not yellow. And similarly for referring to grass browning ( 變黃 ) in autumn, as to my English ear, I would not call the color of grass in autumn "yellow" but rather brown. The book Basic Color Terms: Their Universality and Evolution by Brent Berlin and Paul Kay http://www.amazon.com/Basic-Color-Terms-Universality-Evolution/dp/1575861623/ http://www.amazon.com/Basic-Color-Terms-Universality-Evoluti... was originally published in 1969, and it set the agenda for subsequent studies of how color terms vary across languages. The conclusion (as also related in the blog post kindly submitted here) is that most languages around the world begin with a very basic set of color terms that gradually accumulates more fine distinctions, in a generally invariant order across most world cultures. English has many color terms, but not quite as many BASIC color terms as a few other European languages. We can see this same kind of process in older terms and phrases in English. What is called "red hair" in English (my late dad had that hair color) would more likely be called "orange hair" today, except that there was no word "orange" in English (that comes from the name of a foreign fruit, after all) at that time, so the word "red" had the full semantic range of today's word "red" and today's word "orange." That's quite comparable to the situation in modern Chinese, where there now are competing, very new, words for "brown" ("[coconut] palm-colored" or "coffee-colored" often being used in actual daily conversation) but many historical phrases in which 黃 ("yellow") is used for colors in the same range.

adorable 14y ago

Fascinating article. For a long time I've wondered why we humans usually describe colors as a 3 dimensional space (RGB) where it really only is a 1-dimensional one (wavelength). Took me sometime to realize this was because we have 3 types of rod cells ... and to start wondering how the world looks for the living creatures who have 4, 5 or more rod cells...

bonobo 14y ago
Ask a tetrachromatic woman: http://en.wikipedia.org/wiki/Tetrachromacy#Possibility_of_human_tetrachromats http://en.wikipedia.org/wiki/Tetrachromacy#Possibility_of_hu...
Someone 14y ago
We humans? More 'we computer programmers'. The model 'humans' have more closely follows CIE (http://en.wikipedia.org/wiki/Color_model#CIE_XYZ_color_space http://en.wikipedia.org/wiki/Color_model#CIE_XYZ_color_space). If, instead of measuring it, you asked naive subjects (probing for what they are conscious of), you might even get answers where red and violet are highly separated (as in a rainbow), but I think you can make them question that model easily by creating a palette going from red to violet that does not include orange, yellow, green, or blue. In either case, the problem remains that humans perceive colors that do not physically exist, such as brown (aka 'dark orange-ish') and gray (aka 'shades of white') Historically, some people would have known about mixing paints. That might have change their ideas about color, but that isn't RGB, either. Finally, I am not sure one's color space will change with the addition of types of cone cells (it is cones, not rods that are color-sensitive. It alliterates: kones are for kolor).
andrewcooke 14y ago
i understand what you meant, but, in a sense, if we had "perfect" colour detectors then (ignoring quantum effects) vision would be an "infinite dimensional space" rather than a 1-dimensional one. unless you are restricted to a world in which the light at any one point is monochrome. maybe we would naturally understand fourier analysis in such a world? it's kind of hard to even think about. maybe monochrome light would appear particularly wonderful (or pure) with such eyes? i was thinking about this the other day when i happened to be sitting in the path of some refracted light - it does appear particularly pure even though, with only three detectors, there's no real reason for it to do so (as far as i can see).

bonobo 14y ago

Now our vocabulary grew past the basic white-red-yellow-green-blue-black separation and I believe it will keep growing. I say this because I consider cyan and magenta to be new ramifications of that spectrum, both colors are well recognized by name today. I think this is due to how information is accessible today. Languages seem to be able to separate a hue in two or more once the actual names are already established—when the burden of naming and categorizing hues gives place to the lesser burden of only memorizing the already named hues, then we can afford to create another new name. Although brands of nail polish and ink have dozens of different names for their colors, they often don't reach a consensus, and the same color can have different names between brands. Without a consensus (language is primarily based on consensus) all these subtly different hues can be said to remain unnamed, I guess... I also wonder if there's a limit to this, to how many colors we can name before considering naming all colors in hexadecimal on a daily basis.

crazygringo 14y ago

Funny, in Brazilian Portuguese nobody refers to traffic lights by their colors at all, despite being the same traffic lights used everywhere else in the world -- the light is either "open" (for green) or "closed" (for red). It took me forever to remember to speak in terms of open/closed instead of in terms of colors, when driving...

arturhoo 14y ago
Not quite true. As a Brazilian myself, from the state of Minas Gerais, we most of the times say: the traffic light is green. What is often debated between people from different states is if the light (usually) positioned in the middle is yellow or orange. But yes, sometimes we also say, the traffic light is open/closed.

js2 14y ago

Tangentially related, this reminded me of the color hue test - http://www.xrite.com/custom_page.aspx?pageid=77&lang=pt http://www.xrite.com/custom_page.aspx?pageid=77&lang=pt