Main Page | See live article | Alphabetical index

Hiragana

Japanese writing is divided into four styles: two syllabaries, hiragana (平仮名, literal meaning: "flat/plain kana") and katakana; the Chinese characters known as kanji; and alphabetic letters known as romaji ("Roman characters"). As the alphabet, hiragana can be used for transliteration, a method to represent one language by the alphabet of another language.

Hiragana is used for words for which a suitable Kanji representation is not established, not known to the writer, not expected to be known to the readers, or too formal for the writing purpose. (Previous versions gave an example of "neko", but it is written casually in Kanji or biologically in Katakana.)

To write foreign words, onomatopoeic words, or biological name, katakana is used. However, to give a "cute" appearance, or for text for very young children, hiragana is very often used in place of katakana. It is made of 46 characters, which consist mostly of vowels and vowel-consonant combinations such as "ka" or "hi", but includes one symbol for a lone consonant, which sounds like the English "m" or "n"*. Two diacritics plus the use of digraphss greatly increase the number of possible sounds.

Hiragana developed from man'yo-gana (man'yougana), Chinese characters used exclusively for their pronunciations, a practice which started in the 5th century CE. Literature was written using these characters, and as the forms of the man'yo-gana became simplified (flattened), the hiragana came in to existence, used mainly by women.

Hiragana was not accepted by everyone. Many felt that the language of the educated was still Chinese. However it gained in popularity among women as they were not allowed access to higher education. In fact, The Tale of Genji and other novels were written by female authors using hiragana extensively or exclusively. In modern times, it has become preferred over katakana, which is now relegated to special uses such as borrowed words and names in transliteration.

Most sounds had more than one hiragana. In 1900, the system was simplified so each sound had only one hiragana. Other hiragana are known as hentaigana.

The presence of a percentage of hiragana characters among Chinese characters is usually sufficient to identify a text as Japanese.

If you have a font including Japanese characters, you can view the following hiragana chart:

 
 
base kana ....  
extended with diacritics
  A I U E O  

A

I U E O
 

あ 

い 

う 

え 

お 

  - - - - -
K

か 

き 

く 

け 

こ 

G
S

さ 

*し

す 

せ 

そ 

Z

ざ 

*じ

*ず

ぜ 

ぞ 

T

た 

*ち

*つ

て 

と 

D

だ 

*ぢ

*づ

で 

ど 

N

な 

に 

ぬ 

ね 

の 

 
H

は 

ひ 

*ふ

へ 

ほ 

B

ば 

び 

ぶ 

べ 

ぼ 

  P

ぱ 

ぴ 

ぷ 

ぺ 

ぽ 

M

ま 

み 

む 

め 

も 

*Zi

     (ji) -Zu (zu) - Di (ji) -Du (zu) 

Y

や 

-  

ゆ 

-  

よ 

R

ら 

り 

る 

れ 

ろ 

 
W

わ 

 

 

  

を 

N

ん 

 
 

*Si (shi) - Ti (chi) - Tu (tsu) - Hu (fu)

*There are two symbols for "ji" and two symbols for "zu". The two are NOT interchangeable. The defaults are じ for "ji"; and ず for "zu". The exact spelling rules are referred to as kanazukai. In general, the rules are:

1. ji (hemorrhoids) is written as ぢ.

2. If the first two hiragana of a word are the same, but the second one has a dakuten (see below) diacritic, use the same hiragana as the first one; i.e., ちぢめる for chijimeru.

3. For compound words where the diacritic is added due to compounding, use the original hiragana; i.e., はなぢ hanaji (bloody nose). This does not apply to words where the second element is not considered to be a meaningful, separable element. Thus, even though いなずま inazuma (lightning) is written in kanji as いな (ina) + つま (tsuma), the entire word is considered a single item that cannot be separated, so the default ず (zu) is used rather than づ.

4. Otherwise use the defaults じ for "ji"; and ず for "zu".

5. Note that in Romanization, "di" most generally stands for chi with the diacritic and should be pronounced "ji", thus Meidi = "Meiji". Similarly, "du" should usually be pronounced "zu".

In the table above, if there are two romanizations given for a kana, the former is the Hepburn, and the latter is the Kunrei. If there is only one romanization for a kana, the romanization is the same in both systems.

The Unicode standard also includes a few characters which are not in the standard modern set. These characters: WE (ゐ), WI (ゑ), VU (ゔ) may not be displayable in some web browsers. The "WE" and "WI" kana are obsolete, and the "VU" kana is relatively modern, pronounced as "bwu" to approximate the "v" sound in foreign languages such as English.

The dakuten symbol ゛ is used to indicate that the preceding consonant is voiced, thus shifting k->g, t->d, s->z, and h->b. The handakuten symbol ゜ indicates the plosive, shifting h->p.

The three kana, ゃ, ゅ, and ょ (small "ya", small "yu", and small "yo") change the preceding vowel sound to a glide palatization. They are used after kana in the "i" column (ki, shi, chi, etc.), as follows:
きゃ=kyaきゅ=kyuきょ=kyo
ぎゃ=gyaぎゅ=gyuぎょ=gyo
しゃ=sha/syaしゅ=shu/syuしょ=sho/syo
じゃ=ja/zyaじゅ=ju/zyuじょ=jo/zyo
ちゃ=cha/tyaちゅ=chu/tyuちょ=cho/tyo
にゃ=nyaにゅ=nyuにょ=nyo
ひゃ=hyaひゅ=hyuひょ=hyo
びゃ=byaびゅ=byuびょ=byo
ぴゃ=pyaぴゅ=pyuぴょ=pyo
みゃ=myaみゅ=myuみょ=myo
りゃ=ryaりゅ=ryuりょ=ryo

The "small tsu" (っ) is used to indicate geminate (double) consonants such as the long "k" in the English word "bookkeeper." It is Romanized by doubling the following consonant. For instance, the city of Sapporo would have its name written in hiragana as さっぽろ.
NOTE: The small tsu cannot be used to double the consonants "m" and "n". For example:
おっな cannot be a word; however,
おんな is read "onna" and means "woman".

The special "n" kana (ん) can never be at the beginning of a word. This fact is at the basis of the word game shiritori. (Technically, you CAN have the "n" kana at the beginning of a word, but that only occurs in a few transliterations of non-Japanese words.)
The "n" kana sometimes is directly followed by a vowel kana. This often occurs in the following sort of situation:
恋愛 is read as れんあい (ren-ai)
電圧 is read as でんあつ (den-atsu)

There are ways to get even more sounds out of hiragana, using "small" versions of the five vowel kana. However, this is not really "proper" Japanese, and it is used mainly to make katakana words look cute.

NOTE: If trying to write a Japanese name into kana, remember that you need the original Japanese writing or pronunciation, not an anglicization. English romanization of names are often technically incorrect. For instance, a man may be named Keiichiro (anglicized version), but in kana it is けいいちろう (ke-i-i-chi-ro-u). (Note the final "u" kana.) This problem also goes with names of places: Tokyo is とうきょう (to-u-kyo-u).

BTW: IN Kunrei style, if something like "o" is elongated, you simply add a - (or sometimes a ^) on top of the "o" to notify of its elongation. This is sometimes also done in Hepburn style.

In the image below, both hiragana and katakana are shown (grouped vertically).

Pronounciations

(Vowels are pronounced as in German: "a" is like far without r, "i" is like reed but not long, "u" is like food but not long, "e" is like red, "o" is like cool but not long.

Consonant "g" is pronounced like either "game" or "finger".

"Wo" is pronounced as "o".

"Ha" is pronounced as "wa" if and only if used as a particle.

Collation:

Hiragana are the basis for collation in Japanese indices of names, etc. They are taken in the order shown above (a, i, u, e, o, ka, ki ... re, ro, wa, wo, n). Dictionaries differ in the sequence order for items such as long/short vowel distinction, small tsu and diacritics. As the Japanese do not use word spaces (except for children), there can be no word-by-word collation; all collation is kana-by-kana.