Why Japan has blue traffic lights instead of green | Hacker News Reader

Why Japan has blue traffic lights instead of green

rd.com

192 pointskitebive262 comments

Not a linguist. I feel like this is just an issue of imperfect correspondence between the word "blue" in English and "ao" in Japanese. The article explains the historical reason why ao encompasses both blue and green, so I think the concept of semantic field comes into play here.

As an analogy, a MacBook is a type of laptop, and laptops, desktops and tablets are all IT devices (for lack of a better word). Apple might have you believe that a MacBook is very different from a laptop and belongs in its own category, but to me I would still lump it under laptops. If I was presented with a MacBook, a desktop and a tablet and was asked to pick out the laptop, then it would be clear to me that the MacBook is the correct choice.

Now, midori (green) is a type of ao ("grue"), and ao, kiiro (yellow) and aka (red) are all colours. English speakers argue that green is very different from blue and that they're different colours, but to Japanese speakers ao encompasses midori. If a Japanese speaker was presented with the colours green, yellow and red and was asked to pick out ao (in the context of traffic lights), then it would be clear that green is the correct choice.

There are loads of situations where words in two languages seem to directly correspond to each other, but still they are subtly different especially when the nuances of the words are considered.

I trained as a linguist and you hit the nail on the head.

Word meanings are fuzzy clouds of references and nuances, and every language has slightly different clouds. There is nothing magical about this, despite the recurrent lizard-brain notion that words or names are somehow mystical and intrinsic, and that these differences must somehow be meaningful.

Differences are quite common with colour terms - you don’t need to go to Japanese (blue-green) or Ancient Greek (wine dark sea) for this. My own (European) first language draws a slightly different word cloud around the colours pink and purple than English does, for example. One word is only for hot pink, and the other is for purple and non-hot pinks.

I assure you I see these colours the same as you do. If I were to use the English word “purple” to refer to more of a pink hue, it would be a mere language interference error, not some mystical Saphir-Whorf insight into the culturally-conditioned operation of my retinas.

Words are not perception. This is such a pernicious bit of nonsense, and journalists and writers are especially susceptible to it because it flatters them, in their role as word-smiths. Languages are way more interesting than this pseudo-intellectual mysticism.

The Japanese are just as capable of distinguishing blue and green as anyone else, and they use blue traffic lights for the same reason they drive on the left - because it doesn’t actually matter what convention they pick, so long as everyone agrees on it and sticks to it.

Did you ever tested the Stroop effect in a language, and how it effects depending on the proficiency in that language?

Language is certainly not the sole factor of conscious interpretation of all phenomena, stimuli and mechanisms that induce it, but it definitely is a factor with measurable effects.

And probably this is a skill where individual feel like the largest degree of freedom — the topic of whether this feeling is a mental illusion or backed on hard-wired physics is a distinct point.

Trying to destroy credibility of whole class of people striking them with an anathema like "mysticism" implies forgetting a bit quickly that Descarte’s grand scheme of thought came to live thanks to three dreams, Newton was found of alchemy and Russel dedicated a whole essay specifically to "Mysticism and Logic".

> I assure you I see these colours the same as you do.

Maybe not if I’m colour blind, right?

Now, this is not to promote the extreme other side: I don’t believe in an "absolute relativism" that would allow culture to shape arbitrary anything anyway regardless of any fundamental conditions that enabled human beings to form.

But certainly there a whole set of shade between this two poles (and beyond the linear spectrum they induce).

https://en.wikipedia.org/wiki/Stroop_effect

https://www.cairn.info/qu-est-ce-que-rever--9782749256627-pa...

https://www.philomag.com/articles/le-reve-de-descartes

https://www.gutenberg.org/files/25447/25447-h/25447-h.htm

> it doesn’t actually matter what convention they pick, so long as everyone agrees on it and sticks to it.

That would be a good point, if that actually happened, but Japanese traffic lights have a mixture of green for go and blue for go (with most of them being green).

Can you tell me the location of an actual blue traffic light in Japan?

I went ahead and geoguessr'd one of the images of a "blue" traffic light, and came to the following: https://maps.app.goo.gl/EjYmH34YC19y3fVt9

There is another one to the right as well. They look very blue when turned off, but cyan/green when turned on.

The daylight and the contrast with the trees don't help, either.

That color is common in incandescent green traffic lights around the world, not only in Japan. The idea was to use a blue-green lens to make a mostly green light when paired with a yellow lightbulb

The Japanese specification for the color of traffic lights is certainly written in Japanese. If the specification says the light should have color "ao", then either blue or green or something in between that matches it is OK.

That the color region that word encompasses doesn't match that of the English words "green" doesn't really matter, and is not particularly mysterious.

The article talks about the specification and it's more... specific than that:

> “In 1973, the government mandated through a cabinet order that traffic lights use the bluest shade of green possible—still technically green, but noticeably blue enough to justifiably continue using the ao nomenclature,” Allan Richarz writes for Atlas Obscura.

So there was an intentional decision to adjust the lights to their conceptual preferences. Assuming someone can source this 1973 cabinet order.

In English, things that are orange are often called red. Particularly, but not limited to, hair.

Wait till you find out what they call a blonde person in Croatia :-)

I'm still waiting.

They call it blue.

And orange juice is actually yellow.

Which really blows some people's minds. "Well it's a little bit orange." No it's not, not your regular American bottled or fresh OJ. Straight-up yellow.

(Not denying there are varieties that produce more orange-color juice, like tangerines. But that's not what's on 99% of Americans' breakfast tables.)

It's juice of the fruit-called-orange. How is that not orange juice? We don't call apple juice "red juice" or grape juice "purple juice".

If there's a language in the world where the color and the fruit aren't the same word, I've yet to learn it.

But we do call it red (and white) wine.

> If there's a language in the world where the color and the fruit aren't the same word, I've yet to learn it.

The fruit has the color orange when it's ripe. It's probably one of the most orange things you'll see on nature.

But most people don't even eat it ripe (throwing it away before that point), and the association between the fruit and the color just flies over a lot of people's heads. And yeah, the internals of most of them are yellow.

If the fruit were very unripe, when it has a green rind... I would still call it an orange, not a "green".

It was the name of the fruit before it was the color of the fruit, if I'm not mistaken. Though, I suppose I'm only speculating there. It may have been awhile since I've eaten one, and I prefer the purple Moros anyway, but they've always seemed to be rather orange to me. Quite distinct from the flesh of a lemon, for instance.

>grape juice "purple juice"

That I have seen

No one's disputing that.

What's funny is that most people don't seem to realize that that the color of the juice isn't the color of the rind.

Just look at illustrations of orange juice on Google Images:

https://www.google.com/search?q=orange+juice+illustration&tb...

Most of them are showing an orange liquid that matches the color of the orange rind. It's hilarious. Because somehow, most people don't realize orange juice is yellow, even though they might drink it every morning.

A perverse feedback loop has clearly emerged. It seems that in the US at least, "To further enhance color, manufacturers add up to 10 percent vividly colored mandarin orange juice as well as pigment from orange peels." [1] Further, it seems that juice colour is actually designed and kept consistent. [2]

[1] https://www.americastestkitchen.com/cooksillustrated/how_tos... [2] https://www.xrite.com/fr-fr/blog/beverage-color-control

At least the fruit it comes from is actually orange (and is the origin of the name for the color I believe.)

But if you ever look inside of a fresh North American blueberry, they're pale green! European blueberries might actually be blue inside, I'm not sure.

> European blueberries might actually be blue inside, I'm not sure.

They’re red inside

> Which really blows some people's minds. "Well it's a little bit orange." No it's not, not your regular American bottled or fresh OJ. Straight-up yellow.

Well, it wouldn't be that weird if you were right because we name fruit juices after the fruit, not the juice color, but I’ve never seen orange juice that wasn't distinctly on the orange side of yellow, even if its more yellow than orange, including fresh and bottled American orange juices.

You're just proving my point, of how people insist it's still somehow orange.

Sure, almost nothing is perfectly a 60° hue of yellow. But the color orange is all the way at 30°.

And if you look at the HSL values of the juice in product photos like the following, you'll get hue values of around 52°:

https://www.amazon.com/Tropicana-Orange-Juice-No-Pulp/dp/B07...

https://www.amazon.com/Simply-Orange-Pulp-Juice-Drink/dp/B07...

That's just straight-up part of the band that we call yellow.

For comparison, here's the first result for "banana" in Google Images:

https://www.smithsonianmag.com/science-nature/building-a-bet...

It's 48°, even closer to orange than orange juice. Yet nobody goes around insisting that bananas are "on the orange side of yellow".

It's a case of linguistics somehow trumping what we literally see with our own eyes. It's actually quite astonishing how strong the effect is, even when it's pointed out to you.

That's very interesting. We tend to use a lot of logic as a layer on top of our senses to tell us what we are looking at. One of the most interesting cases (which I can't find a reference for, so sorry) of this that I've come across was an accident in Amsterdam where the person whose car was involved in an accident swore that the other car in the accident had gone against traffic. But from the situation after the accident and eye witnesses it made no sense to believe this. But much later some security cam footage was submitted as evidence that showed the car that had gone against traffic somehow flipping into the position that showed them coming from the right and coming to a rest just like the final scene after the accident was. So, in spite of all defiance of logic the occupants of the first car had it right and the 'eye witnesses' and the accident investigators all had it wrong (and the occupants of the vehicle that caused the accident were lying and could probably not believe they were getting away with it).

I'm sure that in the larger fraction of all accidents eye witnesses will at least get the basics right and that parties are not going to lie about what actually happened. But it's interesting how two minutes after an accident with five witnesses there are probably 10 conflicting stories about 'how it happened' and it makes me wary of any eye witness testimony, especially in less than ideal conditions (night, rain, distance etc). When I first heard about the above case the explanation was that eye witnesses probably heard the crash, looked up and immediately created a mental reconstruction of what must have happened based on what they saw and then reported this as the fact rather than to simply say that they didn't see anything until the moment of impact.

I really should dig a bit more to see if I can turn up something about this case, it is very interesting given your comment because you have provided another example of what I think is the same thing.

> And if you look at the HSL values of the juice in product photos like the following, you'll get hue values of around 52°:

So, Just clicking around your first OJ example and checking lots and lots of pixels, in all the areas of the juice, the juice pixels seem to be mostly around 38-42, with a low of about 32 and a high of about 50.

So, as I said, closer to orange (30) than yellow (60), though also definitely in between.

> For comparison, here's the first result for "banana" in Google Images:

> It's 48°, even closer to orange than orange juice.

Most of the brightly illuminated top seems to be in the 48-52 range, the indirectly-lit side is mostly in the mid 40s, though there are some pixels right along the bottom edge that are also brightly lit (from the reflection from the white surface of the light source above) that hit around 60.

Its not more orange than the juice, for sure.

I don't know what to tell you. I just checked again using Photoshop on my Mac, and the values you're reporting are just not the values.

The cap of the Tropicana bottle has hue values of 25-30°. Which makes sense, because it's clearly visually orange.

So the idea that any of the juice shows 32° is just false. That would be straight-up orange, basically the same color as the cap. But the juice isn't orange. The cap is. In fact, I can't find anywhere on the juice that is anywhere in the 30's, contrary to what you say.

The juice is not closer to orange than yellow. Full stop. It's yellow. Yellow like a banana. That's visually obvious, and it's clear upon exact measurement as well.

(The only thing I can guess is maybe you have some kind of night shift affecting color measurements on your display or something, that's making them redder than they really are?)

> but I’ve never seen orange juice that wasn't distinctly on the orange side of yellow

wait what? I never seen orange juice that was not yellow (in Canada and Brazil)

In Hungarian there is an umbrella term that covers both the color "yellow" and "orange". To further specify it, there is "lemon ..." (citromsárga) and "orange ..." (narancssárga). This umbrella term is "sárga" and it gets translated to "yellow", so "narancssárga" is "orange yellow", and "citromsárga" is "lemon yellow".

So if you were to ask me the color of orange juice in Hungarian, I would reply "sárga", which is true, as it could be either "yellow" or "orange". Many people would still say it is "orange yellow" though, because the name of the color has "orange" (narancs) in it.

Orange juice, like milk, is complicated because it's a suspension and has subsurface scattering, which makes its behavior in different lights very different and much more complicated than "take a picture and check the HSV").

Modern american orange juice looks especially funny because (I thikn) of all the added calcium- it's more of a whitish orange which I find really off-putting.

I recently watched a YouTube video about how "brown" is only a colour because we call it one, but it is really "dark orange" with context.

This one https://youtu.be/wh4aWZRtTwU?si=OZcgqx82umx67VWo probably

That's the one.

I heard it is because the word "orange" is much newer. There were very few things orange in Europe before they saw the tropical fruit orange. Carrots and other things that are orange today were not orange back then.

On Purple vs Pink: it seems that might explain Lego's official name for pieces one might call "pink" use "purple" (with various modifiers) instead.

https://www.thebrickfan.com/wp-content/uploads/2016/05/LEGO-...

Words do seem to have some influence on perception. Specifically, colors appear to be easier to distinguish if they fall within the bounds of separate basic color terms (https://www.pnas.org/doi/10.1073/pnas.0701644104).

I would say appropriately enough this at this level it becomes an issue of semantics, and pragmatics, and specificity,

results like this AFAIK are real (have not been found irreproducible, or limited to e.g. Russian), but, what they demonstrate is best understood in the specific terms of their hypothesis and findings—the "risk" or possible alternative being, to extrapolate the "pernicious" lay belief into some strongish Sapir-Whorf "your reality/my reality."

I.e. measurably faster performance on discrimination tasks, does not map broadly to "perceives differently." Your phrasing was appropriately nuanced, "influences" is a reasonably lay summary...

...but "Japanese-as-first-language speakers perceive blue and green as differently as e.g. English speakers" is still definitely much more true than the alternative.

Personally I was sad to learn strong S-W didn't hold up. It is quite a captivating idea. So too this popular middle ground around color naming.

I find it quite interesting the way experiments like this are construed to try to tease apart linguistic, cognitive, and perceptual factors.

The "fuzzy clouds" argument is both true and important.

But there also are some basic truths about human vision.

We only see three "simplified" different colors, red, blue and green, and other nuances are interpolations our brains make. There is infinitely more frequency information in light that we just don't pick up.

So I would expect that the primary colors red/blue/green, which are grounded in human physiology, were universally recognized across languages. To the extent they're not, that's confusing.

The color receptors in your eyes don’t actually correspond with red, blue, and green with the same wavelengths being able to trigger multiple receptors. On top of this your retina absorbs more blue light as you age.

https://en.wikipedia.org/wiki/Cone_cell#/media/File:Cone-fun...

So, defining the separation between Red, Blue, and Green is really fairly arbitrary.

> Words are not perception. This is such a pernicious bit of nonsense

> pseudo-intellectual mysticism

The influence of words on perception is backed by a lot of research and expertise, and seems apparent on a concrete level: The words I choose affect others perceptions; people who make their living in persuasion (political leaders, opinion leaders, 'influencers', etc) put great effort into chosing words that will influence perception, and they do it to great success.

Why do you think otherwise?

What makes you say otherwise? When you say 'pernicious', that implies negative intent - whose intent? (If that's not meant literally, I take back this particular question.)

I think the original poster is referring to the intent of those that believe in (the strong version of) the Sapir-Whorf hypothesis, which isn't the scientific consensus (at least it wasn't when I studied linguistics/psychology).

As a believe in Hard Whorfism, I think this is rather excellent evidence in support of it...

> I think the original poster is referring to the intent of those that believe in (the strong version of) the Sapir-Whorf hypothesis, which isn't the scientific consensus (at least it wasn't when I studied linguistics/psychology).

When "is not scientific consensus" is considered abstractly, it DOES NOT equal "is false" (if everyone has adequate background knowledge, and is thinking clearly)...but if you observe people (including actual scientists/linguists) discussing concrete instances that involve these abstract notions, then it "almost always" DOES "equal" "is false"...and, if one is to point out the error, it tends to be not well received (rejected due to "pedantry", or one of 5 or so other memes).

Higher in the chain:

> Word meanings are fuzzy clouds of references and nuances, and every language has slightly different clouds. There is nothing magical about this, despite the recurrent lizard-brain notion that words or names are somehow mystical and intrinsic, and that these differences must somehow be meaningful.

Human communication and perception has substantial dependencies on language, and language has a dependency on consciousness, which science does not understand[1] - therefore, pick whatever label you want to attach to this state of affairs, but it is magical/mystical.

> I trained as a linguist and you hit the nail on the head.

At runtime, this tends to render as: ~I know it to be objectively true that you are objectively correct, because I have the necessary specialized training.

> I assure you I see these colours the same as you do

At runtime, this tends to render as: ~It is a fact that I see these colours the same as you do.

> Words are not perception.

This is kinda true - they are (to the degree that they are....which is not known, so the mind conveniently swaps in a simulation, but does not notify us) a fundamental component of perception, but they are not equal to perception - perception has a dependency on words - in general, and on which specific words are used or not) during communication of an idea.

> This is such a pernicious bit of nonsense....

~That which seems like nonsense to me, is(!) nonsense [to everyone].

> ...and journalists and writers are especially susceptible to it because it flatters them, in their role as word-smiths.

~Other people suffer from naive realism and overactive ego, but not me!

> Languages are way more interesting than this pseudo-intellectual mysticism.

~My map of what "mysticism" is is identical to what it actually is (roughly: silly and incorrect "woo woo" would be my guess).

> The Japanese are just as capable of distinguishing blue and green as anyone else, and they use blue traffic lights for the same reason they drive on the left - because it doesn’t actually matter what convention they pick, so long as everyone agrees on it and sticks to it.

This seemingly innocuous claim highlights one of the biggest problems in English, and Western culture: we use the same word (ie; to be) to communicate two ideas with subtle but very importance in meaning: "it is a fact that it is" vs "it is my opinion that it is".

And yes, I appreciate that all this "is" "just" "pedantry", but it also plausibly affects the frequency and severity of war and various other suboptimalities in the world, that in other threads people "assure me"[2] "are" ~"a big deal". Well, if these things actually were a big deal, you'd think people would treat them as such. Sadly, I am very confident that there are people in powerful positions (political "public relations") who understand all of what I say here and much more, likely much better than I understand it.

[1] Notice how I used "understand" with no qualifying terms? This is deliberate, because I have pre-knowledge that this is a scenario where humans get very confused (and often emotionally motivated, depending on the topic, and consciousness is one of those topics) when performing categorization (much of which is sub-perceptual). A soft whorfist might say: This "is" trolling (because that is how it appears to them, and how things appear is how they "are").

[2] it "is" a fact that

Thanks for writing this. I cringe every time I see someone drop the phrase “there’s no direct English translation” as if it means there is some unique insight only able to be expressed in that language. Words aren’t isomorphic between languages, but that doesn’t tell us anything about the range of ideas that are available.

Surely no one believes a native English speaker only experienced schadenfreude after that word was imported from German?

but if some word related to specific concept of culture, do these words exist before people understand the culture?

> I assure you I see these colours the same as you do.

I assure you that you do not. Don't want to get too philosophical or biological on you but we all see colors differently. Every single one of us. Even identical twins.

> Now, midori (green) is a type of ao ("grue"), and..

It isn't. TFA is vague on this, but in modern usage "ao" just means blue. There are a bunch of set phrases where it refers to various other cool colors, but to a modern listener "ao" on its own isn't a category including all those colors, it just means blue.

How do you think this happened? By westernization?

Consider: is "red hair" a color you would consider "red" in a generic context? How about "red cabbage"? Are they the same color?

You’re going to have to define modern usage because words where ao=green are still common in every day language, like seishun or aoba

In set phrases it can refer to things of various colors, but on its own it means blue. Like if you use it to describe a sweater, people will assume you mean a blue sweater.

If those words and others were more obscure I’d agree with your point, but they really are common enough that I think Aoi can mean blue or green in nodern Japanese.

Obscurity isn't involved. In set phrases like you're talking about, the "ao" is basically a category specifier, not a color specifier. Like aoba means young fresh leaves, not necessarily green ones, and aoyasai doesn't only mean green vegetables, etc. There are lots of set phrases like this, both obscure and common, and the color involved can be anything from white to dark gray.

But when ao is used outside of a set phrase, to tell somebody what color you're talking about, it means blue.

As an sort of counterbalancing aside, there are two common japanese words for red, the slightly less well known to foreigners is the name for the color of the sun on the japanese flag: beni/kurenai (a deeper red). Well Chinese has the parallel kanjis too but as far as I understand it for the Chinese the semantic values are flipped: https://www.quora.com/What-is-the-difference-between-%E8%B5%...

Chinese has yet another word for red: 彤. It closer to vermillion or what some people might recognize as "Chinese red" that one sees painted on doors and temples.

I think yellow and amber are a good example of similar confusion in English. According to the NYS driver's manual, traffic lights are yellow and marker lights on vehicles are amber. They both look yellow to me. Doing some research around the Internets, I see people insisting that governments always call traffic lights amber, but that's not true. Either way, nobody knows what "amber" means except people manufacturing lights and people that just looked it up because they're reading a comment about it. But, again, if you ask someone "is the middle traffic light amber?" they'd probably say yes, just like Japanese people will be happy to mark the green light as ao.

Grue is a fascinating concept and it's much more pervasive than Japanese: https://en.m.wikipedia.org/wiki/Blue%E2%80%93green_distincti...

Especially the fact that it tends to impact the speakers ability to perceive shades of the color.

Just a data point that I want to put out there.

I have long heard that Japan has blue traffic lights.

Earlier this year I visited Japan.

Because I had heard about the blue traffic light situation, I paid special attention to the traffic lights.

They all appeared green to me. In both Tokyo and Osaka.

Okay maybe a tinge of blue if you look at it and wonder 'is this blue?', but the only reason I wondered was because I had heard the blue traffic light trivia. If I had never heard about it I wouldn't have given it a second thought.

However, the picture in the linked article seems to have an unmistakable blue color; I don't understand what's going on.

Beautiful country by the way; would love to visit again :)

I am looking out my hotel room window in Hiroshima in this very instant, and I even got my wife to confirm it: The traffic lights definitely are GREEN.

I'm glad you mentioned that you had your wife confirm, because I'm convinced my wife sees more colours than I do, and I can't be the only one. I've had this exact discussion.

    What colour is this?  Blue.
    No,  can't you see it's green?  Maybe it has some green in it
    Okay, so is it blue or green?  Yes!

Why did you ask me then?

Maybe not more colors, but more easily distinguishes them: https://www.psychologytoday.com/intl/blog/brain-babble/20150...

This is different from tetrachromacy, where some women really do see more colors: https://www.optimax.co.uk/blog/tetrachromacy-superhuman-visi...

The mothers of some red-green colorblind children are tetra chromatic, so they actually see more. I think hexa chromatic also exists.

Edit: somehow didn't notice you'd linked this.

I edited it in like under a minute afterwards, you may have opened the page right in the middle. I don't usually bother with an "Edit:" note if it's that fast.

Yes, on average women can perceive more colors than men [1]

Anecdotally, in high school we all took this test for a class [2] and the girls did a lot better than us. Although some guys were able to score pretty well on it. I don’t remember if this was the exact test but it’s very similar.

[1] https://pubmed.ncbi.nlm.nih.gov/21675035/

[2] https://www.xrite.com/hue-test

As a dude who scores perfect on these color tests (this is my superpower), it’s funny on male-dominated places like here and reddit when people discuss logo color changes.

They’re always long threads of very confident people saying how there isn’t even a difference and I’m like lol

Apparently my color perception is as good as I've always thought it was. I scored a 0 on that.

Oh, I’ve taken that test before. I’m a man, but I got a perfect score. I’ve always had a very easy time distinguishing shades and hues of color.

>[2] https://www.xrite.com/hue-test

" Best Score for your Gender -1000000 Worst Score for your Gender 1700045439

About your score: A lower score is better, with ZERO being a perfect score. The circle graph displays the regions of the color spectrum where your hue discrimination is low."

I hate this kind of stuff. WTF is that Best Score?

Hahaha:

> Best Score for your Gender -2147483648

> Worst Score for your Gender 2147483647

It goes from -2147483648 to 2147483647 for me, which I'm interpreting as "our langauge doesn't distinguish signed from unsigned integers, and nor do our developers". Possibly there's also some lack of validation of client-side data involved.

I wonder if there might be a language aspect to this, with males in the US generally having less familiarity with words for the in between colors leading to more difficulty distinguishing between them, with people with more words for greenish shades for example having an easier time seeing different shades of green.

https://boingboing.net/2011/08/12/how-language-affects-color...

I'd have to dig up references, but I understand it's a pretty well documented effect that having words for colors is directly related to a population seeing more distinction for those colors.

If I recall, some early societies didn't have a distinction between green and blue. Given that blue often doesnt show up naturally outside of the sky and it's reflections. And in those populations they would be much less sensitive to distinctions in blue-green.

Which is exactly what the article is about ;) (Well, one example of an early society that didn't distinguish between green a blue.)

For early societies, you might be thinking of the Greeks, though apparently it's somewhat common between languages?[1] For Greek the idea to originate from Homer's writings, with colors being more described in terms of how light or dark they are versus specific hues.[2]

[1]https://en.wikipedia.org/wiki/Basic_Color_Terms

[2]https://en.wikipedia.org/wiki/Wine-dark_sea_(Homer)

Like many siblings, I'm male and got a perfect score as well. I don't think that I have perfect vision, but I'd rather assume the test is flawed.

It's pretty clear that it's "Based on the Farnsworth Munsell 100 Hue Test," and "this is not a replacement for the full test!" I think if you did particularly poorly on the online test then it's worth looking at whether you need to do the full test, as you might have colour blindness (or a terrible monitor). But a perfect score isn't super-meaningful.

Looking at this test on my monitors I would bet it's the 'terrible monitor' before color blindness by a longshot, one shows pretty good color the other is entirely washed out.

What's the scale is not clear. It says 0 is perfect but what about 5?

While comparing it shows typical scores as -(large number) to +(large number).

A lot of that is learned..

However, there are people who are tetrachromats, have cones most sensitive to a fourth frequency of light in their eyes, and those people are also overwhelmingly women.

If I recall correctly there has only been one human actually shown to have tetrachomacy, and the supposition that there are a exceedingly small number of additional ones. I think there was a mechanism proposed that would mean they were all biologically female, but can't remember details.

It certainly seems to be rare enough to have no practical impact (except to perhaps the tetrachromat, who presumably has different metatmers perception etc.)

> those people are also overwhelmingly women.

As far as I understand it takes having two X chromosomes with different versions of the gene that encodes the receptors for green, so only women can be tetrachromats.

I scored a 6, but their lack of data on the distribution of scores makes that pretty hard to interpret.

Does anybody know how to read that?

> https://www.xrite.com/hue-test

Fun test! I'd note for others, the quality of your screen can make a difference - I found one of the lines a lot easier on my phone (first try) than on my computer (second try) - though I got perfect both times.

Your score: 2

Best: segmentation fault (core dumped)

> because I'm convinced my wife sees more colours than I do

A sizeable percentage of women have tetrachromy (5% ? I forgot the exact number but it's not nothing). AFAIR men do not have tetrachromy.

So, basically, some women do actually see more colours than all the men and than most other women.

Maybe your wife has tetrachromic vision? (not sure how it's called)

Please report the exact wavelength for verification.

… in electron-volts to the precision of 10 decimal places.

There's an app for that

Colors are not necessarily a single wavelength, and the human visual system can't distinguish between identical "tristimulus" caused by a single wavelength or a mixture of wavelengths. This is why color monitors with red, green, and blue pixels work. The red, green, and blue are designed to stimulate your long, medium, and short wavelength receptors. Models like CIE1931 map wavelengths to tristimulus values, which your monitor then maps to the right intensity of its red, green, and blue. (Display technology is, of course, imperfect. You can see redder reds, greener greens, and bluer blues than your monitor can emit, though newer monitors with Adobe RGB or DCI-P3 have redder reds. It's also possible that the green light from your monitor is not perfectly stimulating only your medium wavelength receptors.)

Some colors that look very real cannot be a single wavelength of light, like magenta. That color is a figment of our imaginations. (Rather, there is no wavelength of light that stimulates your long wave receptors and short wave receptors without stimulating the medium wave receptors. But, magenta does, because it's actually a blue wavelength and a red wavelength.)

Stop ruining my jokes with facts!

My memory form my time living in Japan is that they do not have a consistent shade of blue/green that their traffic lights conform to. You will see completely different colors depending on the location, and even see different colors within the same location. Some areas are more blue and some areas are more green.

I have never seen the big main lights using blue instead of green anywhere in Japan; the traffic lights for pedestrians however were blue in some places.

you would know if you read the article :)

Are you sure? This seems to be the full article:

> www.rd.com

> Checking if the site connection is secure

> www.rd.com needs to review the security of your connection before proceeding.

weird, maybe you have to send them your blood, fingerprints, and a list of your favorite laundry detergents in order to get through?

From the Guidelines:

>Please don't comment on whether someone read an article.

Which you'd know if you read the guidelines (:

https://news.ycombinator.com/newsguidelines.html

I have been to Japan a couple of times. The first time I was there in 2016 many of the lights were very blue with hint of green. I noticed when I was there over the summer that the lights were getting greener, many of them pretty much just green. I think they have been transitioning, perhaps as they have updated to LED lights or something they have been making them more green?

I wouldn’t be surprised if it had something to do with the Tokyo olympics. Are other cities also green now?

Yes, I was in several other cities and noticed it throughout. Spent half of the summer in Sapporo (my favorite Japanese city so far!) and it was definitely noticeable there.

It depends on where you are in Japan, it's mostly Green but there are some Blue ones occasionally.

And yet another data point: there are places in France with blue lights, instead of green.

How old are you roughly? The reason I ask is that the eye lens yellows with age and transmits less blue and violet through. So colors that would look blue-green to a younger eye would look more green to an older one.

The "go" signal in traffic lights in Japan is 青い (あおい) "aoi".

That adjective in Japanese means both blue and green. The Japanese language didn't have a widely-adopted word for just "green" - 緑 (みどり) "midori" - until after WWII.

Green apples or vegetables are still "aoi", for example. I lived in Japan for four years and never saw a blue traffic light.

Do you know more about the history of this color use? It’s surprising to me that midori uses the kanji for green in Chinese if it’s such a recent idea in Japan.

What I was told, when I asked about this, was that "midori" existed as a Japanese word but was not in common usage. That is what Mizutani-san from JAL Academy told me back in 2009 anyway.

EDIT: I apologize for the vague sourcing!

Green (綠) in Chinese also came later. In Old Chinese, 青 was generally used to represent both blue and green colors.

While the word 綠 to mean green has been attested as far back as 1000 BC, the idea that it was a separate color rather than describing a shade of 青 is relatively more recent. Wikipedia[0] indicates that it was adopted in the early 20th century in Chinese (as part of vernacular language reforms) and after WWII in Japanese, though these claims are currently marked with [citation needed]. While both are relatively recent, the usage in Chinese did have a longer period of time to take hold.

[0] https://en.wikipedia.org/wiki/Blue%E2%80%93green_distinction...

I don’t know about the green kanji, but the traditional “Jueju” Poem by Chinese poet Du Fu has the line “山青花欲然”, and “青” the Kanji for Blue in Japan means green in this context, as it describes “山” which means mountain.

By the way there’s a variety of kanji that reads “aoi” in modern Japanese. 青い, 蒼い, 碧い are all “aoi”, but we use 青い mainly as blue today.

Oh that’s interesting. Yeah 青 is used in Chinese as another word for green. I don’t know that it has any connotations of blue, but I did grow up in the US, so perhaps there’s nuance to that character I’m unaware of, or the character slightly changed meaning when made a kanji.

This is pretty far off base - for a start, the first JP law on traffic lights in the 1930s used midori, not ao. Per jp references, ao/midori split into distinct meanings around the Heian~Kamakura periods (so 800-1000 years ago).

> for a start, the first JP law on traffic lights in the 1930s used midori, not ao.

Have a source for that? This section[1] of the ja.wikipedia article could apparently use your revisions - both kanji are acceptable.

[1] https://ja.wikipedia.org/wiki/%E4%BA%A4%E9%80%9A%E4%BF%A1%E5...

I don't follow. Are you suggesting the link you posted contradicts something I said?

If you want a source that the 1930 law used midori, see here, under 発祥期:

https://ja.m.wikipedia.org/wiki/%E6%97%A5%E6%9C%AC%E3%81%AE%...

Are you referring to reference number 184? I'm looking for a primary source for your statement that midori was the colour standardized in law since the 1930s, because the section I linked about colours explicitly calls out ao and notes the difference in different jurisdictions.

> your statement that midori was the colour standardized in law since the 1930s

In 1930, not since the 1930s. The 1930 law used midori, but it was changed to ao in 1947 after that became the more popular usage. The page you linked was about the current law.

A lot of traffic lights in the US are actually "not just green".

The traffic lights have an abnormally high amount of blue light for being an overall green light, to help people with red-green colorblindness to tell the difference between the red and "green" light.

If you take a look at the green light on a dash cam, you might see that the color is abnormally bluer than what you might see in real life.

Japan has green traffic lights, there are no two ways around it. I’ve lived in Japan for 3 years in all the major cities. They are green, just like everywhere else. They just call them blue.

Why they do that is something you can argue about and I’ve heard Japanese try and come up with all sorts of explanations:

“The character for blue is easier to learn than the one for green, so it’s better to say blue with children.”

“They used to be blue, so we still just call it that”

“Green isn’t an original Japanese color. Before foreigners arrived we didn’t call anything green”

“They are blue!” - usually these are people who will call traffic lights in Europe and smarties blue even when speaking English.

It’s just a quirk. I think Japan places the border between blue and green different from most western cultures and that gives.

The Reader's Digest article is just a mediocre summary of the Atlas Obscura one it links to: https://www.atlasobscura.com/articles/japan-green-traffic-li...

Although the latter does contain a complete misrepresentation of what "grue" means.

The explanation is still unsatisfying. Japan had green lights, referred to in law by a word which means blue but has traditionally encompassed green as well. Pedants pointed out that it would be better if the law explicitly said green. Rather than changing the law or letting things be, the government ordered lights be changed to blue-green. This is, on the face of it, completely insane behaviour, but neither article attempts to explain why the government did this.

I just scanned a few articles on this. TFA and its source don't mention this, but the first JP traffic law for traffic lights (from 1930) used midori/green, following US standards. But apparently newspapers mostly used "ao" for whatever reason, and that was the usage that became widespread. Then later on (1947) the traffic law was updated to say "ao", presumably to match common usage. Of course the actual lights were green the whole time.

Then in the 1970s the rules changed to recommend bluer shades of green, and new lights made afterwards reflected that. None of the articles I found gave a concrete reason; a few suggested bluer lights were easier to see, and a few implied the change was motivated by the "ao" name.

> This is, on the face of it, completely insane behaviour, but neither article attempts to explain why the government did this.

How is it insane? Did the outcome turn out to be chaos and mass murder? Are the Japanese people all becoming lunatics due to the cognitive burden of trying to remember the differences between one foreign color word and the other? Will the Reticulan Space Navy destroy the Japanese for their insults to chromaticity?

This isn't insane. It's just boringly bureaucratic.

This reminds me of when I took my drivers' license test in Japan -- they'll do an eye exam including whether you can recognize the colors. The first color was this weird mix between blue and green, and when the administrator asked me what color was showing, it took me a moment to try to find a word to describe the color... then I realized, I'm at the DMV, they want "ao", so that's what I said.

I passed that with flying colors by saying グリーン

Other people have covered the linguistic aspect enough so no need for me to rehash it, but, the picture in that article is far more blue than any traffic light I've seen in Tokyo in my months there. If I had to describe them with a single word, I'd choose green without hesitation - they just have a bit of a blue tinge to them.

Some example photos from google image that look closest to what I've seen: https://www.pond5.com/stock-footage/item/129881591-traffic-l... https://depositphotos.com/editorial/traffic-lights-ginza-dis...

I've not really left Kantō so maybe they are more likely to be more prominently blue elsewhere? Kansai and Hokkaidō google image searches show (mostly) green traffic lights too.

The article seems to imply outside of Tokyo with "Elsewhere around the island" and perhaps "Drive around Japan long enough". It would be nice if the article was more specific here, but it's also just a Reader's Digest summary.

I've seen almost-blue stop lights (go lights?) in the USA, and before everything was LED the color variation between various lights was pretty wide. You could almost always figure it out in context, but absolutely speaking some red lights were almost as yellow as some other light's yellow.

I’ve seen very blue traffic lights in Japan both in Kyoto and Tokyo. I think the weirdest part of this story that isn’t talked about in the article is that Japan doesn’t seem to have any regulation or consistency defining which shade of blue/green the traffic light should be. Different areas will have different colors and even the same area will have different colors.

Color blindness compatible design is, IMO, a very neglected part of the equality debate, in many other fields. This disability has a very skewed gender ratio due to how it works genetically.

Traffic lights are still fine for colour-blind people aren't they? The order of the lights doesn't change.

Night time?

Also lights with more than three bulbs, some indicating left or right turns, of which will take on two or more colours themselves. Some lights have a bulb above red for cyclists and transit. Some lights called beacons are a single bulb that take on several colours.

https://www.ontario.ca/document/official-mto-drivers-handboo...

Yes. They can still drive fine, but don't get the benefits of color coding, like being able to infer whether red or green is lit from the light reflected off of other objects. That becomes somewhat important in low visibility situations where the position is hard to determine.

It seems the harder difficulty is red and yellow, see the comments in this article, but led lights have improved the situation for some:

https://www.color-blindness.com/2007/02/06/colorblind-at-the...

There are a few different types of colorblindness, so I wouldn't extrapolate that.

For me personally, the green light looks much closer to a white than to, say, the green of a grass.

I just happen to have a friend with red-green blindness, who was my reference point in answering. It's good to point out there are different types with different impacts, though.

Funny thing, I'm NOT colorblind and apparently I only rely on the color. I've been driving for over 10 years now, mostly in the capital that has traffic lights everywhere, and sitting at my desk now not looking at a traffic light, I wouldn't want to bet money on the order.

I always assumed red was at the top as it's the most important. It's the first one to peak out behind tall trucks.

Some places have the lights sideways, and usually red is on the left, but it's not always consistent.

If I were designing them I'd always have the red be noticeably larger.

It depends on their kind of colorblindness.

I see green lights as white, which is fine most of the time because of the order.

The exception is when there's a lot of glare from the sun setting behind me, and I sometimes can't tell if it's green, or just the sun reflecting at the right angle. It also washes out the red and yellow, making it harder to see if they're lit.

I am the same--green traffic lights are nearly white. Normally, it's enough to know it's not the red or yellow light. Sometimes at night--face with an array of white lights in the distance--it can be an issue.

At my last eye test at the RMV, the clerk tried to tell me I wasn't color blind. I mean, sure, I've been able to drive with it for 40 years, but I know damn well I'm color blind.

Horizontal lights are not consistent across countries. The linked article has an image of a Japanese horizontal traffic light with red on the right, whereas an American traffic light would have red on the left.

And some countries don't have horizontal lights at all.

https://youtu.be/sXbHdKJ1D78

I don't think I've ever seen horizontal lights in Europe at all. Fairly common in Japan though.

Which is even worse. I grew up in an area with only vertical lights. After several years of driving I ended up in an area with a horizontal light, one end was white and I had no clue if that meant stop or go. (I'm colorblind so I normally think top=stop, bottom=go)

Most color blind have issue to see shades and subtle differences. They see red and green just fine.

The people who do not see red at all however, see red as black. So, they do not see red traffic light at all.

> The order of the lights doesn't change.

For 3 lights streetlights no. For 2 lights streetlights it depends. They can be (in DE) either red and yellow or yellow or green.

AFAIK yellow + green is only ever used for turning traffic and is always installed together with a regular 3-bulb traffic light for the other directions. Stand-alone two-bulb traffic lights are always red + yellow, with red on top, just as usual.

With the switch from incandescent bulbs to LED traffic lights it would be very easy to change the shape of the lights: e.g. octagonal for stop, triangular for caution, and circular for go. Keep the colors, but add shape as well. Seems like an obvious and useful change to me.

That might fall apart for people with poor visual acuity.

If they're roughly the same size and in the same order as the round lights today, people with poor visual acuity would still see more or less what they see currently.

If you can tell neither the color of the light, nor its shape...

Should you be driving?

Almost fifty years ago, I worked with a guy who was color-blind. This was in Colorado, where at the time some of the little towns in the mountains still had traffic lights arranged horizontally rather than vertically. He must have known the order at one time--I think I learned it for my license test--but of course with little chance for reinforcement tended to forget. I suppose that such lights are all gone now.

They're common in at least four states still: https://www.youtube.com/watch?v=sXbHdKJ1D78

> The order of the lights doesn't change.

That argument has two issues: First, not all US traffic lights have order, as single-light ones replace stop signs in some (mostly suburban/rural) areas. Second, light position is still hard to discern at night.

My personal take: the red and yellow hues in traffic lights are often indistinguishable for those of us with red-green colorblindness.

I learned this lesson in my first ever public lecture. Approximately 10% of my audience had some form of colour blindness, rendering the majority of my figures incapable of communicating their intent.

Since then I have rendered all figures for public consumption in black and white, and lines instead of surfaces where possible.

Depending on what exactly your figures were/are, you might benefit from the viridis colour map, available in matplotlib and R (and probably in many other places). It's colourful, so those without colour blindness get a richer display than if you used greyscale, and the colours chosen aren't too affected by most common forms of colour blindness. But for those that really can't see the colours, even if you literally just convert to monochrome, it still works. It also looks nice :-)

Here's a talk by the creators: https://www.youtube.com/watch?v=xAoljeRJ3lU

Here's a little article about it (with some R specifics): https://cran.r-project.org/web/packages/viridis/vignettes/in...

Am I wrong that in that article magma appears to do better? Particularly in the last two (red-blind and grayscale), it seems like viridis looks the same for large parts of the scale, while magma retains its distinction.

I think it just won the poll [1] because it subjectively looks nicer. NJS, in that video, said something like: I don't care which wins, they're all so much better than the current situation. But magma is a good choice too and it's also available in matplotlib.

[1] http://bids.github.io/colormap/

Android has a colorblind developer option in the settings you can use to quickly check how it might look for others.

I just turn that on, open the camera and look at my color pallete through the camera.

I have the relatively typical male, minor colorblindness. Using color as an signal is fine, but just make sure there's something else that's independently conveying the info.

Also, another way to find out how common colorblindness is among your male friends: go rock climbing in a gym. Color is used on the holds to signify the route, and colorblind people who quickly start asking about color as they're climbing.

Most green lights in the US have a blue tint so that color blind people can see the blue. Most but not all, which is really frustrating as a colorblind person. More than once I've been sitting at a light until my wife told me it was green as there is one light near me without that blue tint. (we are making a left turn so light has both the round red stop light and the green left turn arrow on at the same time)

There is no clicking/hum that distinguishes those 2 states? If that's the case, you should probably check with council to have those nearby lights changed into disability-friendly setup. You may be surprised how effective it may be, maybe council has some part of the budget allocated to this so they will give it higher priority (or not, but at least try, you can help other people too)

Crosswalk lights in the US may have audio but usually not the roadway lights, and depending on the jurisdiction the crosswalk may not always get the light at the same time as the roadway, not to mention that you may not hear it if you are at the centermost lane of a road four or five lanes wide in each direction.

There is for pedestrians, but when I'm in my car with the windows closed I can't hear them.

I'm pretty sure the international agreement on road signaling completely forbids any stop and go light from being on the same place, so that you can distinguish them without depending on any color.

But well, the US never follows any agreement it signs, so yeah, that happens.

Sure, but green is so washed out for me that the green position may not be there at all for all the good it goes. If you tell me the green is lit I'll look close and see it is on, but I will not notice without being told to look.

Red cabbage or Blue Cabagge, in northern Germany is also called red cabbage (Rotkohl), same color for some regions of Austria and Switzerland (Rotkraut).

In southern Germany it is called Blue Cabagge (Blaukraut).

If you ask people in Germany what the color of the vegetable is, they will answer "purple" (lila). There are some strange ways in the evolution of a language, depending on the region and events in the region.

It's normal, whatever normal means in this case, to think that a color is the same as a similar hue, in my example above, between red and blue you can find purple, violet "lilac" hues.

As a personal anecdote, the name for (orange) carrot in Southern Bavarian dialect is Gelberübe, Rübe for root vegetable and Gelb for yellow, also yellow is connected with orange in the brain and the language of the people.

A purple carrot has the honor to be called "lilane GelbeRübe", or purple yellow vegetable root in English.

Red cabbage is a bit of a weird example because its colour depends first on the acidity/basicity of the soil it was grown in, and then when you're using it on the acidity of other ingredients.

For example, if you cook it with apple (which is acidic), it will turn redder.

I don't think this distinction stems from the same source. Blue/green distinction is a clear step in linguistic development that happened in different stages for different civilisations, but they're more of a result of linguistic developments, local available colours, and manufacturing capability than a reference to plants.

This particular cabbage is coloured by a chemical that responds to the pH of the stuff it comes into contact with. The colour can range from quite bright red to quite clear blue, and even green or yellow.

It's perfectly possible for the red cabbages to be turned into blue dishes, and the pH of the soil will also have a large effect on the colour produced by the plant. You can see on various stock photos how the plant has a clear blue hue (before harvesting, at least): https://www.alamy.com/stock-photo/rotkraut.html?pseudoid=562... Even ignoring the leaves, the outside of the parts that generally get cooked have a clear blue hue in many pictures I can see.

The reason "lila" wasn't used to describe the cabbage is that the German language lacked a word for it. It entered the German vocabulary somewhere in the 19th century (https://en.wiktionary.org/wiki/lila#German). In a similar fashion, there was no separate word for orange in many European languages until somewhere around the 15th-16th century, when oranges (the fruit) were starting to get imported. You had yellow-red and other hue combinations, but it wasn't as separate as it is today.

I'm sure there were people who used "Rotblaukohl", but it makes sense that only one colour remained.

As for the carrot, orange carrots were actually not all the common for centuries. The original plants now known as carrots were imported to Europe from the middle east and cultivated in the Netherlands, but orange carrots weren't all that common in Europe before the 16th/17th century. The base plant of the orange carrot was actually white/yellow and got its orange outside hue quite some time later, after selective breeding. I wouldn't be surprised if the carrots that were first exported to modern German areas were still yellow in colour. Gelbrübe for orange carrots makes a lot of sense, historically.

I do like the "purple yellow root" name, especially since the first carrot cultivars to reach Europe (long before orange/yellow carrots) were actually purple. I don't think they received quite the popularity carrots received, at least not much further north than the Mediterranean.

>The reason "lila" wasn't used to describe the cabbage is that the German language lacked a word for it.

This is totally it, some people just decided to name it the way they saw it, and depending on the region were they lived.

Rotblaukohl was probably the middle ground, but there's in German a rule to join two colors: das blaurote Kleid (the blue-red dress), in this case it would mean the dress has two colors blue and red.

I might be very far from my area of expertise TBH.

Carrots were originally white/yellow. The Dutch made an orange variant at some point.

If you read any Chinese/Japanese literature, for example Buddhist or zen texts, it is very common to encounter very weird uses of the words blue or green-- because they mean things that somebody used to these distinct categories will classify nature according to these boxes. Trees are green, bodies of water blue -- even if that is more complex than that, some pines can be very blue, rivers or lakes can be much greener than blue, etc.

Some translators avoid this issue by always using the term "blue/green", which is really awkward, and I couldn't find any explanation for it before learning about "the crayola-ification of the world" [0]. Before I thought that was a poetic literary device.

It is really hard not to think of the Sapir–Whorf hypothesis when learning about this. At the same time, taoist philosophy tends to point out that misery comes from our mind's discriminating eye; everything is categorized in boxes, good and bad, concept and not-concept. Maybe having more categories is better from a technical standpoint, but more difficult to handle from a spiritual standpoint? At the same time, this spiritual view tend to see man as needing to overcome his beastly nature, and thus this added technical discrimination is not burden since it is simply part of the path towards a higher level of consciousness?

[0] https://empiricalzeal.com/2012/06/05/the-crayola-fication-of...

Do you have any examples of such Zen texts?

I can remember it was a case in a translation of Dogen's works, which may not be the most accessible in the first place.

> because they mean things that somebody used to these distinct categories will classify nature according to these boxes.

Very interesting. Could you explain this part in a little more detail?

I'm not sure I'll answer properly but I'm thinking of how information retrieval works in the brain and how similar it is to word vectors and heatmaps.

There is a well known trick (I don't know if there's actual research behind) that when asked for a tool and a color, people will answer red hammer. There are many tools and colors but these come to mind quickly because they are so frequent, simple, etc. Therefore the concept used for information retrieval implicitly creates a set of all possible words that satisfy it. For instance "bird" will make the person think of pigeons, sparrows, crows, so it naturally implies "flight". It's only by precising either "flightless" or a specific flightless bird that the association is removed. The implication goes both ways: flightless birds tend to not come to mind, despite chickens being extremely common. Furthermore, it is quite counterintuitive to just take arbitrary conjunction of categories (e.g. a bird or a chair). By comparison, discriminating further is very easy, and people tend to be able to much easier think of different elements of the same subcategory (e.g. different breeds of pigeons).

Koans tend to revolve around erroneous thought or language patterns, and so having the "blue or green" category was an obvious example of falling into either of my known boxes (blue or green) before being reminded that the concept encompassed both.

Although the color of traffic lights is determined to be green, there is a reason why they are called green lights in Japan. The green color used in traffic lights in Japan is the closest to blue (blue-green) within the range of CIE regulations in consideration of people with color blindness, and historically, the green color used in Japanese traffic lights is One reason is that the range of "blue" is wide.

As a British person living in the USA I often enjoy that the line between blue and green shades definitely differs between cultures and there are certain shades of teal that US/UK folks will strongly disagree on.

I’ve since noticed that I think this question of where “that is definitely green” ends and “that is obviously blue” begins is different for lots of countries

I also find it fascinating that cultures have so many words for colours or so few words for colours either way it's interesting.

Here is a good video on colour and how many cultures perceive colour and name them, or not.

"The surprising pattern behind color names around the world" https://www.youtube.com/watch?v=gMqZR3pqMjg&t=35s

> Hundreds of years ago, the Japanese language included words for only four basic colors: black, white, red, and blue. If you wanted to describe something green, you’d use the word for blue—“ao”—and that system worked well enough until roughly the end of the first millennium, when the word “midori” (originally meaning “sprout”) began showing up in writing to describe what we know as green.

There's something going on here that the article doesn't explain.

The new word is written 緑, which is also [the Japanese simplified spelling of] the Chinese word that means green in specific. But the Chinese word has always referred to a color, and it's never meant "sprout" -- the shuowen jiezi, written 800 years before the period the article indicates, defines it "帛靑黃色也"!

So by the time this word is being written down, there is no sense of the concept "sprout" at all.

The article seems to present this as a case of conceptual innovation internal to Japanese, which would have been much cooler than the apparent reality of the Japanese starting to use an already well-established concept more often.

Now I want to know why, if midori originally meant "sprout", it does not seem to have the spelling 芽 ("sprout" - 萌芽也) as a possibility...?

That explanation seems wrong to me, when you refer to sprout like things i.e. spring leaves on trees you use 青葉 - blue leaves.

https://kotobank-jp.translate.goog/word/%E9%9D%92%E8%91%89-2...

As far as I can tell from Wiktionary (https://en.wiktionary.org/wiki/%E7%B7%91#Japanese), though I don't speak Japanese, the Japanese word midori has had the meaning "bud" or "sprout", but the Chinese words written with the same character don't. Spelling and word identity are distinct. This confuses me sometimes when looking up Japanese words.

> Spelling and word identity are distinct.

What I'm saying is:

1. If "midori" began with the meaning "sprout", then...

2. its spelling in early texts should have been 芽 [sprout] and not 綠 [green].

The spellings were fixed a long time before the article identifies the change as taking place.

The Japanese spelling of a word must develop by either (1) the Japanese borrowing a Chinese word with the meaning of the Japanese word; (2) option #1, but the Japanese meaning of the word later shifts; (3) option #1, but the Japanese spelling of the word later shifts [this would be weird]; (4) indigenous innovation of a character; or (5) refusal to use a character at all.

"Midori" appears to have begun its life by being spelled as if it meant "green" and to have continued to mean "green" since that time. This is strange if it originally meant "sprout", and I'd like to know more about the claim and the history.

Ah, I see, I misread your comment quite badly. I don't have any references to look this up in, but perhaps the ancestor of the word midori is first attested in a syllabic spelling in Old Japanese with only the meaning "sprout", and when it came to be spelled logographically as 綠, it was mainly used with the meaning "green". But the question could be asked in the Tea Room of Wiktionary, where it would probably find its way to someone more knowledgeable than me.

They speak of this as if it was a Japan-only thing.

The blue-green distinction is a late feature of most languages.

Even Greek and Latin did not have as clear a distinction as we have today.

> They speak of this as if it was a Japan-only thing.

Which other countries have blue traffic lights?

In China, for example, the word for green 綠 is used for the light

綠 is used in Japanese for the color of green tea, but they use 青 for the green light

There's also 翠 in Japanese to mean a bright green, which in Chinese also has a blue connotation (based on the color of the kingfisher who dives 淬 to catch fish)

I'm not doubting that multiple languages have that aspect to it, but as far as I know that does not result in blue traffic lights. I have not heard of an instance of a blue traffic light in China to go with your Chinese example.

Because they use 绿, not 青 so it would not result in blue lights

But in Korea they have the same confusion:

https://www.dreamstime.com/traffic-lights-road-blue-speed-li...

To the best of my knowledge it's different in Korea. As far as I know the call it blue lights but they are green.

They are not blue, but because the word for blue is also ambiguous they are more like a teal color

This discussion is weird, because the Kanji 青 in Japanese was borrowed from classical Chinese (presumably during the Tang dynasty around the 7th to 9th century).

So using Chinese as an example of "another language" having green-blue confusion isn't very meaningful since modern Chinese aka Mandarin also inherits pretty much the same 青 albeit popular usage differs somewhat.

Yet the fact that there's no blue traffic light in China just hinges on the fact that people don't commonly refer to the "green" in traffic lights as 青 in China (but you could). The word 青 is used to refer to a broad spectrum of green and blue in different contexts, just happens people generally refer to the traffic lights as "綠" instead. And since there's no ambiguity to "綠" there's no confusion of whether the traffic lights should be "green" or "blue", and thus you don't see blue traffic in China (hopefully).

The word "ao" in Old Japanese already had the blue/green confusion before the character 青 was borrowed to write it, so if classical Chinese did as well I guess that would be two such languages.

> The word "ao" in Old Japanese already had the blue/green confusion before the character 青 was borrowed to write it

What would the evidence for this be? It's clear that the word "ao" does not derive from the Chinese pronunciation of 青. But its meaning could easily have shifted under the influence of the spelling. Do we have a source that writes the word without spelling it 青?

This? https://ja.wikipedia.org/wiki/青#やまと言葉の「あお」

The US. They are mostly green, but a lot of them have a blue tint for the colorblind. I suspect other countries as well, colorblindness is very common and this is a simple adaptation.

I wouldn't call "green with a slight blue tint" a blue light, personally. That's still a green light.

As someone who has trouble seeing green, often I would call them blue. I don't expect my experience is normal.

viridis / caeruleum

but those don't exactly map to what we call green and blue today.

That's the same in Vietnamese as well, it's "xanh" for both. You have to add some extra precision if you really want to differentiate them.

what's interesting is that the chinese characters used for blue and green ended up being backwards between China and Japan

How do you come, that they had no distinction between blue and green?

The Greek had no word for blue, which is the reason Homer described the sky with different words. But I never heard that the sky was green-isch.

While on the other side the Greek had more similarities between green and yellow.

https://duckduckgo.com/?q=green+sky&t=ffab&iar=images&iax=im...

The Greek said the sea was green, for example; so did the Romans.

The Romans did have a word for blue, but it only applied to the sky.

The Romans used the word "viridis" both for green and for blue-green.

Nevertheless, when they wanted to describe more precisely a color, for example when Pliny the Elder had to explain the difference in color between emeralds and beryls, they used expressions like "green like the leaves" or "green like the grass" for green, and "green like the littoral sea" for blue-green.

> The Greek said the sea was green, for example; so did the Romans.

This isn't exactly a surprise; they were both heavily involved with the sea.

English speakers talk about the sea being blue, but that's not because it is blue by the standards of English speakers. It's because most people don't bother to give any thought to the things they say. If you go look at the sea, it is obviously green.

Which sea?

Although Homer speaks of the "wine-dark sea" which is an odd visual picture for us.

The fact that Greek did not have a word for blue is a myth without any ground.

Already in Homer the word for blue was "cuaneos", which means of the color of "cuanos", like "chruseos" (golden) means of the color of "chrusos" (gold).

"Cuanos" was initially the name for the blue pigment that now is named "ultramarine blue", which was an expensive pigment imported from Afghanistan.

Later, "cuanos" was also used as a name for other cheaper blue pigments that could be used to substitute the expensive ultramarine blue, i.e. for the azurite mined in Cyprus and for the artificial pigment "Egyptian blue".

The English word "cyan" comes from the Greek word "cuaneos", but due to a misunderstanding it is used now for blue-green, despite the fact that it was never used for blue-green by the Greeks. In the Ancient Greece and Rome, when blue-green had to be distinguished from green, it was specified as the color of the beryls, or as the color of turquoise, or as the color of the littoral sea.

While in Greek there was an unambiguous word for blue, what was missing was a word for green. Green is mentioned very rarely in what I have read, and when it is mentioned they use one of the following expressions: the color of the emeralds (smaragdinos), the color of grass (poodes) or the color of leek (praseos).

In the early Greek authors, "chloros" that is now used to mean green in many scientific terms was not used for green, but perhaps for yellow or yellow-green, e.g. Homer uses "chloros" for the color of some honey.

https://en.wikipedia.org/wiki/Blue%E2%80%93green_distinction...

Yeah and they also don't realise that "one word for two colours" is just nonsensical.

All words are descriptions for a continuous range of colours. Oa apparently just covers a particularly huge range. But you could equally say "ha those stupid English people only have one word for lilac and purple. Idiots!"

I think the most common real-world example is Russian and Italian, which have a specific word for light-blue which is distinct from the word for blue.

I assume this is like how ‘pink’ is a special word for ‘light-red’.

You see the same thing in other areas. In Mandarin Chinese, the sounds represented by "si" in English vision or "r" in English virile are the same sound.

An English speaker (like speakers of many other languages) would find this ridiculous, but it is true that the two sounds represent either side of a completely arbitrary threshold applied to a continuous phenomenon.

It's like in English the sound represented by "ś" in Polish "ściągać" and "sz" in Polish "szumieć" are the same sound (usually written "sh") :).

If you go out and buy LEDs for use in a traffic signal in North America, you will have to specify the colour as "cyan". That is because the monochromatic wavelength sits in the place between green and blue. So the same sort of situation in naming occurs in that part of the world too. You can call the colour green, but you could just as legitimately call the colour blue. It really just ends up being a convention.

One thing I discovered when buying some custom plates made from old traffic signals was that while the glass for red and yellow were obviously those colors, the glass for “green” was actually blue.

They appear green because of the warmer light that incandescent bulbs give off. I imagine that things transition to LED they’ll choose warmer light frequencies to continue this effect? https://mytrafficlights.com/why-does-the-green-lens-on-my-tr...

In Salamanca (Spain) there are some old traffic lights that they have blue (not a full blue more or less a half step between green and blue) color too.

I think that they turned to blue for things of years and years of sun rays and raining and other climate things.

I sell traffic lights. What you consider a “green” light in the USA is actually called “blue green”. I believe the reason for this is increased visibility for older drivers.

It’s for red/green colorblind people. Obviously you don’t want someone who can’t tell the difference between red and green while driving.

Maybe Korea also had blue traffic lights in the past. But they still call green traffic lights 'blue' even though all of them use a green coloured light.

I think the use of blue to mean both blue and green has also been present in Korea for a long time.

I wonder where it came from? In my limited experience if Korea and Japan share something culturally it often has its common derivation in Chineese tradition.

Both come from the Chinese term 青 which in modern Chinese would by itself refer to colors anywhere between sky blue and grassy green

青天 - clear skies

青草 - green grass

清葡萄 - 청포도 - white grape

As vague as 青 might be, I'd have to allow that it's a more accurate description of the color of a grape than "white" is.

To this day I insist that American traffic lights are red, orange and green. I know what yellow looks like and that is not yellow, it is orange. As a child, I couldn't understand the concept of calling it yellow despite it emitting orange light. Now, I understand the concept of colloquially referring to things by incorrect names, but I still hate it.

In New Zealand the middle light is (or at least was) called amber.

Now I live in the US and the lights here look the same as NZ but I'm pretty badly colorblind so they're just words to me :)

Horizontal rather than vertical lights were new to me here. They make sense because they don't swing around in the wind.

The concept of "red = stop" and "yellow = slow" existed prior to electricity and artificial light: in rail transportation, a yellow flag meant proceed with caution.

When traffic lights were first created, they simply didn't have the technology to create a great pure yellow, so amber was used instead, and people understood the color they were aiming for.

Makes sense, since Yield signs and similar warning signs are all yellow as well. But being born into a world where everyone calls an orange light yellow is confusing nonetheless!

In France, it's called orange.

I used to wonder, as a kid, why most traffic lights in my region of Germany were red, orange-yellow and obviously turquoise. I remember drawing traffic lights (I loved drawing road layouts when I was young) without any use for the green pen. Nowadays, they all seem to be very much green, at least the new LED ones.

I’d say here in the US the LEDs skew much more blue than the incandeswnts they replace. Almost a teal.

They are green. Japanese people just call them blue (青). The colors we call green and blue are just ranges on a spectrum. Where you draw the line between them to call one green and the other blue is partially a matter of personal perception and maybe more a matter of culture but there is no absolute right answer.

I'm not sure if there's any connection to why Japanese tend to lean towards blue rather than green but the Japanese word for green, midori (緑) is linguistically different from most other colors. It doesn't have a true adjective form and is instead used as a noun or "adjectival noun". Does this difference create a subconscious aversion to it in the Japanese psyche? I really have no idea.

We also call traffic lights “green” in North America and Europe, but if you look carefully they’re actually in the “cyan” color family, much closer to turquoise or aquamarine than a green.

https://en.wikipedia.org/wiki/Web_colors

As mentioned, aoi ringo (blue apple) is a common term. When I moved to Japan many years ago, on my first visit to a McDonald's, I was initially excited then let down by the prospect of getting a "blue apple" milkshake.

I believe that Greek, too, lacks a word for blue. Modern Greek speakers just use the English word blue as "mple" where "mp" stands in for the B sound which has no distinct letter in modern Greek.

Surprised me, and I wondered what sound Beta signifies in modern Greek. Apparently /v/.

In Modern Greek, the voiced plosives of Ancient Greek, i.e. beta, delta and gamma, have become voiced fricatives, while the aspirated plosives of Ancient Greek have become voiceless fricatives.

To deal with the voiced plosives that are present in many modern loanwords, from languages like English, they had to develop ways to write them, using the existing letters of their alphabet, and the solution was to use digraphs, like "mp" for English "b".

Thank you.

How do modern Greek speakers pronounce the ancient words? With the new sounds, or the old sounds? Are there many ancient words in modern Greek?

I do not know if this remains true nowadays, but in the past most Modern Greek speakers pronounced the ancient words with the modern pronunciation, because this is how it was taught in schools, despite the fact that the Ancient Greek pronunciation was very different.

This is the same like with the Latin language used by the Catholic Church, which traditionally had a modern pronunciation very unlike Ancient Latin, or how the Latin was traditionally taught in English schools, also with a modern pronunciation unlike the ancient pronunciation.

In the last half of century, in many places the teaching of ancient languages has switched to using pronunciations as close as possible to what is known about the ancient pronunciations, but I do not know if this has also happened in Greece.

The Modern Greek has replaced many ancient words with more recent loanwords, but as a reaction to this they have created a form of the language, Katharevousa, which was purged from many loanwords and which used a high proportion of Ancient Greek words. Katharevousa was used for many literary works in the past and it was the official language of Greece since independence until 1976. I do not know if there is still much use of Katharevousa for anything, except for reading the older books and other publications that have used it.

Thank you! As a speaker of another ancient language, this is very interesting.

https://en.wikipedia.org/wiki/Blue%E2%80%93green_distinction...

Color in general can be subjective, even between people of the same culture and language group. There's an interesting experiment with color in this video around yellow lasers that might help get the idea across.

https://youtu.be/XQnmMjYHgcM

There's an amazing book on the naming of colors and more: Through the Language Glass: Why the World Looks Different in Other Languages by Guy Deutscher.

It's a nice article and has lots of details.

But I searched my photo history for Japan and here are the first few that have traffic lights in them: https://imgur.com/a/S815y7o

This should help you determine what colour the lights are in your perception. Some are taken with a Canon DSLR and some are taken with a Pixel 3.

Nice photos!

Thank you :)

Common, not a single mention of Sapir-Whorf Hypothesis?

https://en.wikipedia.org/wiki/Linguistic_relativity

https://www.verywellmind.com/the-sapir-whorf-hypothesis-7565...

No, because it is discredited pseudo-science.

Even taking this as premise, it doesn’t mean it should not be mentioned.

Knowledge that some hypothesis where disproved is also crucially important to understating and formation of sounder models.

Regarding the adequacy of the assertion itself, every reader should judge by themselves to which level it extends.

> Even then, midori was considered a shade of ao. As you can imagine, this sudden switch-over had lasting effects in Japan. One of the most popular forms of Japanese wordplay is numeric substitution—writing numerals in place of words or letters with the same phonetic value.

Mad that they just add the numbers thing there, but never mention if there's a number play with green/blue

https://archive.ph/uj03N (rd.com requires JS)

I don't know if its a urban legend or not, but I always heard that this was the upstream cause this plugin [1] in jenkins being a day one install on teams I was on

[1] https://plugins.jenkins.io/greenballs/

Yep.

https://www.jenkins.io/blog/2012/03/13/why-does-jenkins-have...

I still don't forgive Jenkins for giving in and replacing blue with green.

TL;DR: The traditional Japanese language did not have separate words for green and blue; instead, it used a single term "青" (ao) to cover both colours. Over time, the language evolved, but the initial categorisation of green as a type of blue stuck around in certain contexts, like traffic signals. This linguistic trait led to the designation of traffic signals that are green in colour as "青信号" (ao shingou), which literally translates to "blue signal."

This comment section is full of interesting topics and comments, thank you very much! I'm glad this story started such a wonderful discussion!

It’s interesting (and unsurprising) that regions that are geographically isolated (islands) are more likely to have unusual road and traffic characteristics.

Shame the Readers Digest has gone for such a modern bland looking site, and dropped their classic logo.

Alaska has blue traffic lights sometimes as well

as other comments have pointed out, this also happens with other east asian languages. vietnamese is another one.

Reply on news.ycombinator.com