Posts Tagged ‘Kanji’

Adapting Kanji to Japanese

February 4, 2010

As I discussed in my last post, Japan imported writing from China.  However, these symbols – called kanji in Japanese – are morpheme-based, representing both sound and meaning.  Thus kanji could at first only be used for writing Chinese.  In time, however, a system for writing Japanese eventually emerged.

Adaptation began with Man’yōgana.  Named after the first anthology of Japanese poetry, the 8th century “Man’yōshū,” Man’yōgana are kanji used purely for their pronunciation, with no regard for meaning.  They made it possible to write in Japanese, but they weren’t codified; to write the sound “a” for example, any character pronounced “a” could be used.   Given that kanji number in the tens of thousands, this made for a cumbersome system.  Nonetheless, it was a start, and Man’yōgana in turn gave birth to the two syllabic alphabets still used in writing Japanese today.

One of these syllabaries is katakana (like the other types of writing I’m talking about, the name refers both to the syllabary as a whole and to the individual symbols).  Katakana were originally used for writing notes or pronunciation glosses in the spaces between columns of text.  This meant that katakana’s symbols had to be simple, in order to fit into small spaces legibly.  This chart from Wikipedia shows how katakana were created.

In each cell katakana is on the left and kanji is on the right, with the parts used to create the katakana colored red.

The other syllabary is hiragana.  Hiragana was the direct evolution of Man’yōgana, created not by taking pieces of kanji, but by writing whole characters in a simplified cursive form.  Furthermore, hiragana were used not for annotation, but for general writing.  This chart from Wikipedia shows the evolution of hiragana.

In each cell the original kanji is on top, the intermediate cursive form is in the middle in red, and the modern hiragana is on the bottom.

Hiragana replaced Man’yōgana as the means of writing Japanese-language poetry and literature, but Chinese continued to be used for official documents.  However, in these Chinese documents, there had to be a way of writing the names of Japanese people and places.  Man’yōgana was one method, i.e. kanji could be used to spell out names.  The other method was to use kanji for meaning instead of sound, and to then assign these kanji the native pronunciations.  This approach became increasingly common, with more and more kanji acquiring a Japanese reading in addition to the original Chinese reading.  In cases where it was unclear how a word was meant to be read, a gloss could be written next to it in katakana.

As a body of native literature was written in hiragana, people got used to reading and writing in Japanese.  And as it became more common to apply native Japanese readings to kanji, using katakana to annotate text when necessary, it became possible to write Japanese using these meaningful kanji.

The system of writing that finally emerged combines kanji with both hiragana and katakana.  Kanji are used to write most nouns, verb stems, and adjective stems.  Katakana are used for foreign words and as the equivalent of italics.  Hiragana are used for most everything else, notably for inflectional endings for verbs and adjectives, for grammatical particles, and for words with no kanji or whose kanji is obscure.  Hiragana also replaced katakana as the script generally used for pronunciation glosses.  And in modern Japanese, Arabic numerals are common and Roman letters are used for SI units and for a few words like “CD.”

Here’s an example sentence.

Kanji is in orange, hiragana is in blue, and katakana is in green.  A word by word translation would be  “Interesting program [subject marker] few because, television [topic marker] much watch-[negative verb ending].”  Fully translated, it means “There aren’t many interesting programs, so I don’t watch much TV.”

In this sentence, katakana is used for the loan word, “TV.”   Kanji is used for the native noun, for the stems of the adjectives, and for the stem of the verb.  Hiragana is used for the grammatical particles, inflectional endings, and for the words “because” and “much.”

I’ll stop there for now.  More on kanji and hiragana in my next post.


December 17, 2009

Japan had no written language until regular contact with China began in the 6th century.  Over the next several hundred years, Japan adopted many aspects of Chinese culture, and one of these aspects was writing.  At first, though, all of this writing was actually done in Chinese, not Japanese.  Why?  Because Chinese isn’t written with an alphabet.

Chinese is written using glyphs known as hanzi, meaning “Han characters” – “Han” being the name of China’s dominant ethnic group.  In Japanese, this word is pronounced kanji.

Kanji began many thousands of years ago as pictograms.  So the word “bird,” for example, was a picture of a bird.  But what if you want to write “hawk,” or “duck,” or “heron”?  And then what about different varieties of hawk, duck, and heron?  Trying to have a different pictogram for every bird quickly becomes ridiculous.  And what about abstract ideas?  Pictograms are too cumbersome, and so most kanji were instead created by combining other kanji.

One type of combination deals purely with meaning.  For example, 明, a character that means “bright,” is composed of 日 and 月, the characters for “sun” and “moon.”  This category is small, though.

The majority of kanji were created such that one part indicates pronunciation and another indicates meaning.  Take 松, the character for “pine tree.”  The left half comes from 木, meaning “tree,” and the right half is 公, which means “public.”  A pine, therefore, is a tree whose name is pronounced like the word for “public” – in ancient Chinese, that is.

However, it’s generally not the case that kanji equal words.  Rather, most kanji correspond to morphemes, the units of meaning that make up words.  For example, the English word “geology” is composed of two morphemes, geo-, meaning “Earth,” and -logy, meaning “study.”  So if we used kanji to write English, that would be a two-character word.  (As it happens, “geology” is written with three kanji – 地質学 – which mean “ground-qualities-study.”)

Oh, and when I say that 松 was made by combining 木 and 公, note that the creation process happened millennia ago; you don’t make up new characters every time you write a sentence.  Sometimes artists and authors will invent new kanji, but it’s not something that is normally done.

So.  Each kanji character corresponds to a morpheme (usually), and each kanji is either a pictogram – stylized beyond recognition – or a composite of pictograms, most often consisting of a part that hints at meaning and a part related to pronunciation.  It’s complicated, but not arbitrary.  But what happens if you want to use kanji to write a different language?

Well for one, the clues to pronunciation are immediately rendered useless.  “Pine” and “public” may be pronounced the same in Chinese, but they sure aren’t the same in English.  And what do you do if one language has a morpheme that the other doesn’t?  And how do you write names?  “Kevin” means “kind and gentle,” but if you decide to use those characters to write “Kevin,” then you’re assigning them two separate pronunciations.  And in that case, how does anyone know that they’re reading the name and not the sentence fragment?

All of these problems meant that when Chinese writing was introduced to Japan, it could only be used for writing Chinese.  This was better than having no writing, but understandably, people wanted to find a way to write in their native Japanese.  Eventually they did, although it took centuries and left the Japanese language permanently changed in the process.

But I’ll talk about that another time.

Xi’an: Day 2

May 31, 2008

The 7.9-magnitude earthquake that hit Sichuan province has killed more than 68,000 people by the current count and left millions homeless. I was fortunate enough to be well out of the country when the disaster occurred, although in any case the closest I got to the future epicenter was Xi’an, more than 400 miles distant.

View Map

But earthquakes certainly are scary things. The quality of a building’s construction is probably the biggest factor determining safety, but even if your building is earthquake-proof, that won’t matter if a mountain falls on you, and even if you don’t have that danger, things like gas leaks and soil liquefaction can still do you in.

And aside from loss of life, there’s destruction to homes, businesses, and infrastructure. The 1995 Kobe Earthquake “only” killed about 6,400 people, but it still holds the record for the costliest natural disaster to befall a single country, with $200 billion in damage. Kobe was Japan’s busiest port before the quake, but although the city has recovered, it’s now the number four port and isn’t likely to regain its former prominence.

But back to my travels.

On my second day in Xi’an, I visited the famous Terracotta Army, which lies a fair distance outside the city. Statues were discovered in 1974 by farmers digging a well and are still being slowly excavated and restored. The life-sized figures number in the thousands and the enclosures protecting the several excavation sites are as big as aircraft hangers.

The army was built to accompany the First Emperor into the afterlife. The Qin dynasty also soon followed him into death, but the emperor definitely left his mark on history. As he forged the empire, he standardized measures, currency, laws, and writing. He undertook immense canal and road building projects, and he connected several walls left over from the Warring States period into the predecessor of the Great Wall. He also ruthlessly repressed Confucianism, which stresses that rulers should be virtuous and benevolent, burning books and burying hundreds of Confucian scholars alive (Confucianism was revived in the next dynasty, though). He did everything on a grand scale and his burial arrangements were no exception.

The Terracotta Army includes not only infantry, but also archers, charioteers, horses, officers, generals, and even some jugglers. Their uniforms differ by rank, but even among soldiers of the same type, you can see variation in faces and clothing.

I think this is a so-called “heavenly horse,” a Ferghana.

The horses were imported from Bactria, and although they’re much smaller than Arabians, they’re larger and faster than the little Mongolian horses that had previously been used in China and they were highly prized.


After returning to the city, I spent the end of the day at the “Stele Forest.”

The Stele Forest is a collection of more than 2,000 stelae, kept within a Confucian temple. The stone blocks are inscribed with all manner of things, including books, poetry, memorials, and the occasional map or diagram.

There’s some pretty major stuff in the collection. For example, this stele is the first record of Christianity in China.

It was carved in 781 and it documents the spread of Nestorian Christianity since it entered the empire in 635. Christianity was later outlawed and the stele was buried and forgotten until being unearthed in the 17th century.

Content aside, on many stelae, the writing itself is a treasure. The work of a number of famous calligraphers is in the collection and a variety of styles are represented. Alas, due to glare, I don’t have any decent pictures of the best calligraphy, but I can at least show some different types of writing.

This is seal script, so called because today it’s mostly only used in official seals.

The newly codified characters mandated by the Qin emperor were written in seal script. Although it is thoroughly abstract and stylized, it’s closer to the writing system’s pictographic origins than later forms. I think this stele is a guide to writing seal script made after it had passed out of daily use, because it’s subtitled in the “regular script” seen here.

Regular script, which evolved out of clerical hand, is what people use today. It’s the easiest style to read and write.

And for an even bigger contrast, this is cursive, known as grass script.

I find it amazing that anyone can actually read grass script. When I see Japanese written in the style I can never make out more than a couple words.


Well this entry took forever. I’ll try to speed up my pace as far as time allows, because at this rate, I’ll be back in the States before I finish writing about China. Just two months to go. Yikes.



Here’s a neat chart from Wikipedia showing how the character for “tiger” evolved from a pictograph to its modern, abstract form.

As an aside, it’s worth noting that only a small fraction of Chinese characters have direct pictographic origins. Most are composed of one part indicating the general meaning and another part supplying the pronunciation (at the time the character was invented, that is). For example, “language,” 語, is made of a part indicating speech-related words, 言, plus a part giving the pronunciation, 吾, whose own meaning (“my”) is unrelated to that of final character. Alas, things are complicated by how pronunciation has changed over the centuries. Although 語 and 吾 were both pronounced something like “go” during the Han dynasty, modern Mandarin says “yǔ” and “wú” respectively. So while there is logic in how characters were made, the task of learning them today is mostly rote memorization.