EDIT: Started a bounty with hope of getting more definitive and elaborate answers, e.g. timeline of when color names started being used in Japan.

In Japanese language, there are colors that are i-adjectives: 青い, 赤い, 黒い, 白い etc.

Then there are colors that are original nouns: 緑 {みどり}, 紫 {むらさき} etc.

And finally color nouns that are made of [something]-色: 茶色 {ちゃいろ}, 黄色 {きいろ} , 灰色 {はいいろ}, 桃色 {ももいろ}, 銀色 {ぎんいろ} etc.

Why are there different word classes for colors in Japanese? Is it because they have different historical backgrounds and etymologies?

  • 1
    黄色 is read as きいろ. But 黄色 is a little different from other words ending with -色, because 黄 itself is the name of a color whereas 茶色, 灰色 and so on mean the color of tea, ash, and so on. Jul 29, 2011 at 13:25
  • @Tsuyoshi. Fixed. Thanks. Also I noticed that 黄色 is a な-adjective unlike the other 色's which are nouns.
    – Lukman
    Jul 29, 2011 at 13:43
  • 2
    And there is also i-adjective 黄色い! I wonder what's the significance of the yellow color in Japanese ...
    – Lukman
    Jul 29, 2011 at 14:43
  • 1
    And there are colors from English, written in katakana. Apr 7, 2019 at 8:05

3 Answers 3


Surprisingly, I wasn't able to find much information on this outside Wikipedia, even though it's popular trivia in Japan, but here's data from a book about color names:

In China circa 1000 BCE, the only kanji evidenced is white. By 820 BCE, this grows to white, yellow, orange-red, and green-blue-black. In 770, 赤 is no longer used for orange. In 750, 青 is invented for green-blue, and by 720 緑 is invented for green and 黒 for pitch black.

So much for the use of kanji, which presumably had strong influence on Japan when they were imported in the first millennium AD. What about Japan's native words? According to Wikipedia, they are the following:

Aka(赤)which shares an origin with "akarusi" (bright), "ake-/aku" (open), "akaaka to" (brilliantly/flaming) etc.

Kuro(黒)which shares an origin with "kurasi" (dark), "kure-/kuru" (darken/end) etc.

Siro(白)which shares an origin with "sirusi" (mark), "siru" (knowledge), "sirasira to" (obvious speech). Apparently the original word is Old Japanese "sirusi".

Awo(青)which shares an origin with "awi", the plant indigo, and which is claimed to be the antonym of siro.

Therefore, it is claimed, ancient Japanese had four colors: light-warm, dark-cool, distinct, and indistinct. In the transition to medieval Japanese, light became red, dark became black, distinct became white, and indistinct became blue-green. These comprise the four i-adjective colors.

The origin of midori is unknown.

Later, as another commentator said, "iro" was added to words to mean "the appearance of tea" etc., and later still color words were borrowed outright from English. 黄色, yellow, comes from the appearance of sprouts.

For more information, Wikipedia has some book recommendations at the bottom of its color names page.

  • Aw, dude! That book is sweet! Aug 7, 2011 at 2:48
  • Wow thanks for this revelation! I didn't expect to find a culture that once had red/orange vs black and white vs blue/green. This post deserved the +50 :)
    – Lukman
    Aug 7, 2011 at 4:56
  • 10 years later, still an amazing answer.
    – Sam
    Oct 12, 2021 at 22:42

Let's talk about い-adjectives: 暗いペンを作ってる。 In pre-kanji Japanese language, adjectives had an い ending. They always end with い、あ、お、or う before the い suffix. They never end with え, as in せ, れ, て, or け. The reason there are many い-adjectives is because they were inherent to Japanese language, even before Japanese people began using Chinese characters in writing. Arguably, there was no written language before the Golden Scrolls came to Japan, though. We have to base our assumptions on word of mouth, decryption of artifacts, and early writing in Japan, so it's difficult to be sure. Anyway, い-adjectives are used for abstract and concrete concepts because the words were originally い-adjectives and not derived from nouns, like の and な adjectives both were.

Let's talk about な-adjectives: 紫なペンを作ってる。 な-adjectives are words that were added to the Japanese language as nouns. な was added to the end of nouns which seemed to have a use as an adjective. Na-adjectives can have あ、い、う、え、or お before the な suffix. People never use な with colors, as in 紫なペン, except when they're using their own vocabulary. For example, if you have a desk full of different kinds of purple pens, you could ask someone if they liked 紫な; however, nobody would say that. People always say こんな紫のペン. The reason you don't see any な-adjective colors is because colors are extremely simple to understand--objective/concrete--unlike い-adjectives which embody both abstract and concrete concepts, and の-adjectives, which are for concrete concepts, only. な-adjectives are used to describe abstract/subjective things, only.

And about の-adjectives: グレーのペンを作ってる。 They're a lot like な-adjectives, but they're for concrete/objective concepts, only. It's possible to think of な- and の-adjectives with colors: こんな色->colors like this and この色->this color. I think that some linguists call na-adjectives 'adjectival nouns' because they are "derived from nouns", but な-adjectives are actually derived from の-adjectives, which are derived from "nouns". Perhaps it's confusion with the possessive form: あなたのペン->your pen? If you say 紫のペン->a purple pen, purple is definitely an adjective in translation. 'Purple's pen' wouldn't make sense, in translation. Many color words are の-adjectives. の-adjectives are thought of in the spirit of the possessive tense, though, in that the adjective's qualities are contained in a specific object or set of objects.

Combining our knowledge of い-, な-, and の-adjectives, we can say: only い- and の-adjectives exist for colors. So, we know that colors are thought of as being concrete concept.

Dialects: 茶色 means tea-colored. Since the birth of kanji, Japanese people have been combining kanji to form meaning. In English, we use a hyphen to indicate an adverb, and an adverb can take an adjective's place, becoming an adverbial adjective. In Japanese, there are so many compound words, and we have suffixes indicating part of speech, so it would be over-productive to hyphenate.

Besides, kanji is more complicated than the English parts of speech accommodate. For example, 馬鹿 means horse-deer (originally an idiom). Because it's an idiom, we have to think of the relationship between the words to understand the meaning. It's from ancient Chinese history. One of the emperors of China never wanted his subjects to undermine him, so he had all of his advisers come to him for a test. The emperor called a horse a deer in front of his advisers. He killed each adviser who corrected him. People thought that this strategy of divining an adviser's competence was stupid, and an idiom was developed in China: 'horse-deer'--"as stupid as calling a horse a deer" or "as stupid as correcting the emperor who calls a horse a deer", perhaps. It's very complicated.

Words like 茶色 (chairo) and 馬鹿な (bakana) adapted into the Japanese language, despite their complexity. Because Japanese language was influenced by the native あいぬ, immigrant Koreans, and immigrant Chinese, there were many options and dialects; however, 茶色 and 馬鹿な are a standard because, for some reason or another, people accepted the dialect.

色 history, from the birth of kanji in Japan: Why do I say "since the birth of kanji"? Because 色 is one of the kanji included in the Golden Scrolls that were initially sent to Japan, from China. The first record of Chinese characters being used in any significant way, in Japan, was during the time when the Golden Scrolls were sent to Japan. The Golden Scrolls were some of the first documents sent to Japan. It's not the first document, but it's very close to the top of the list. When the Golden Scrolls came to Japan, the Chinese basically gave Japan their first definitive writing system.

I think that い-adjectives and な-adjectives likely existed prior to の-adjectives because I have been told that many times. History teachers and text books have told me that な-adjectives were used in conjunction with nouns to form adjectives, and with foreign words that were abstract and fit better as adjectives than as nouns. の-adjectives were a later distinction.... If you actually look at Japanese writing, though, they pretty much all come at the same time (right when the initial phonetic alphabet, man'yōgana, came into place). At this point, Japanese people were using all three kinds of adjectives. Perhaps the people who wrote the books did some kind of comparison of man'yōgana to older Japanese texts, in an attempt to get more in touch with their history, and this is why people are taught that adjectives came about in this order?

In terms of history, who can be sure, though? The pronunciation of Japanese words, even beyond 500 AD, isn't clear to anybody. This is because there was no commonly used phonetic alphabet, and because so many dialects were borrowed. There was a lot of immigration to Japan in the early days of Japanese writing. Japan had many local dialects, as well. It wasn't until much later that a unified Japanese language was formed. The creators of today's official Japanese language appreciated language, much as we do today, for an art form. That, combined with the simplicity of seeing 色 and knowing it's a color, is probably why they chose to use words like 茶色 rather than some pre-existing い-adjective. Now, if there ever was an い-adjective for 茶色, it's deprecated.

Anyway, the kanji 色 first came to Japan in 57 AD. 色 was originally used like 表情 with 顔色, and it means: 'the expression on the face'. Other than that, it's used with color. That's the purpose of the kanji..., to identify color, as counters indicate something you're counting. The kanji is supposed to be two lovers sitting next to each other, looking at each other, and reading each others emotions. So, 色 is like: 'you can tell if you look closely'.

Great Seal Scroll characters came to Japan as Han-period Chinese writing (206 BCE–220 CE). It looks different today, but 色 has basically been there since the beginning of written language, and therefore historical records of language, in Japan. You can basically assume that it's always been used with colors. I know for sure that it was used prior to 622 AD in association with colors, in authentic Japanese, because the librarian at our school showed us the kanji being used in a texts. 色 just happened to be used, which was exciting for us. The book explained how Chinese writing was hard to discern from authentic Japanese writing, but this was one of the ways....

The writer of the poem, which we couldn't actually understand, appears to be either Japanese, or some eccentric Chinese person who wasn't using Chinese properly, due to the syntax. There's some hentai kanbun. Maybe it's still a Chinese person, trying to describe something, but it seems like a bunch of Chinese characters that don't belong together. For example, if I were to say something in English like: "It's super-cala-fragil-ist-ic." There are too many word parts, and it looks very odd.

  • 3
    1. Ancient adjectives end with , not . 2. 紫なペン is unnatural. 3. There is no such thing as の-adjectives. If it is follwed by , then it is a noun.
    – user458
    Aug 3, 2011 at 19:01
  • 1
    Some dictionaries have a classification for "adj-no" though.
    – Flaw
    Aug 4, 2011 at 0:21
  • 1
    @sawa, There's no such thing as 'adjective' in Japanese. It's really a subjective argument. People break the word after the suffix 'no' and call it a no-adjective to contrast it with a 'na' adjective using the same root word. This makes it a single word, and to correlate it with an English equivalent, we have to give it some kind of label. 病気の人 and 病気な人 both use adjectives with the root word 病気, but they have a different suffix, の and な. I suppose you could say the の means something other than '~ened' vs '~ly' or the adj 'sick' vs 'sickly' or 'sick-like', but it would remain subjective. Aug 6, 2011 at 14:52
  • 1
    Can you tell me more about き? Aug 7, 2011 at 2:22
  • 1
    I like how someone always has the right dictionary... and everyone else has the wrong dictionary. ;)
    – summea
    Feb 28, 2012 at 20:25

There is a study by Berlin and Kay that colors are scaled from the basic ones to the advanced ones. The more basic, the more likely that it will have a mono-morphemic word representing it in a language. There is also an implication relation: 'If language A has a mono-morphemic color term W, then it will also have mono-morphemic words for the more basic colors.' What you found follows this scale. Colors like red, black, white are basic colors and a mono-morphemic word representing them exist in many languages. Green and purple are more advanced, brown, gray, pink, silver are even advanced. That indicates that at some old time, Japanese only had the colors that correspond to the i-adjectives. Later, some more colors were added, and even later, no new color words were added but people just used some things with that color to refer to the color.

Edit: Sorry. Blue was not a basic color. The overall tendency of the generalization is correct, but there are some exceptions.

  • Where can I read further on the study?
    – Lukman
    Jul 29, 2011 at 14:20
  • @Lukman There is a good wikipedia article. You can expand from that. en.wikipedia.org/wiki/…
    – user458
    Jul 29, 2011 at 14:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .