Hiragana

Hiragana (ひらがな) is one of the two Japanese syllabaries. Each sign writes one short sound, usually a consonant plus a vowel: か is ka, の is no, す is su. The standard chart is called the gojūon, 'fifty sounds', though modern Japanese uses 46 basic hiragana. It is arranged as a grid: five vowel columns あ い う え お (a, i, u, e, o) and ten consonant rows, k, s, t, n, h, m, y, r and w, with ん (n) on its own at the end. Every Japanese dictionary, phone contact list and class register sorts in this order, so learning the chart means learning how Japan alphabetises.

A few cells do not sound the way the grid suggests. し is shi, not si; ち is chi; つ is tsu; ふ is fu. The y row has only や ゆ よ (ya, yu, yo) and the w row only わ and を; the gaps in the grid above are real gaps, not missing characters. を is used only as the object particle and is pronounced o: ほんをよむ 'I read a book'. Two other particles are also said differently from their spelling: は as the topic marker is wa (こんにちは konnichiwa), and へ meaning 'to' is e.

There is an older order too. The iroha poem (いろはにほへと…) uses every kana of its day exactly once, ゐ and ゑ included, and was the sorting order until the gojūon took over in the Meiji period. Its first syllables still label things the way a, b, c do in English: seats, exam answers and the parts of a musical score.

Dakuten, handakuten and small kana: how hiragana writes every sound

Two small marks multiply the chart. The dakuten ゛, two short strokes at the top right, voices a consonant: か ka becomes が ga, さ sa ざ za, た ta だ da, は ha ば ba. The handakuten ゜, a small circle, turns h into p: は ha becomes ぱ pa. That gives 20 voiced kana and 5 p kana, all on this page as ready-made characters, so you never need to type the mark separately. ぢ and づ sound the same as じ and ず and appear mainly where two words join: はなぢ hanaji 'nosebleed', つづく tsuzuku 'to be continued'. ゔ (vu) writes a foreign v sound in hiragana; it is rare, since loanwords normally take katakana ヴ.

Small kana change the kana next to them. Small ゃ ゅ ょ fuse with the i-column kana before them into one syllable: きゃ kya, しゅ shu, ちょ cho, whereas full-size きや is two syllables, ki-ya. Small っ is not pronounced on its own: it doubles the consonant that follows, and the difference matters: きって kitte is a postage stamp, きて kite is 'come'. At the end of a word in casual text, っ marks an abrupt stop, as in あっ 'oh!'.

The other small kana are specialist. Small ぁ ぃ ぅ ぇ ぉ show a glide or a drawn-out vowel in song lyrics, dialect writing and manga: すごぉい. Small ゎ survives in a few classical spellings (くゎ kwa). Small ゕ and ゖ are variants of the counter ヶ in place names and dates: 三ゕ月 'three months'. The small 𛅐 𛅑 𛅒 (wi, we, wo) and small 𛄲 (ko) were added for transcribing Ainu and for old texts; they need a recent font and are the last ones a learner needs.

Where hiragana comes from: from kanji to onnade to the 46 of today

Hiragana grew out of Chinese characters. In the 8th century Japanese was written with kanji used only for their sounds, a system called man'yōgana after the poetry anthology Man'yōshū. Writing those characters fast in cursive script wore them down into simple curves, and by the 9th century, in the early Heian court, the worn shapes had become a script of their own. The origins are still visible: あ is a cursive 安, い is 以, う is 宇, か is 加, の is 乃, and を is 遠.

Men of the court wrote official documents in Chinese, so the new script was first taken up by women writing diaries, letters and poems, and it was called onnade, 'women's hand'. The great prose of the Heian period was written in it: Murasaki Shikibu's Tale of Genji and Sei Shōnagon's Pillow Book, both around the year 1000. The name hiragana (平仮名), 'plain kana', came later.

For centuries every sound had several hiragana shapes, because different kanji had been worn down into the same syllable; a writer chose by taste and by what came before. In 1900 Japan's Ministry of Education fixed one shape per sound for schools, and the others became hentaigana (変体仮名), 'variant kana', listed in the last groups on this page. The spelling reform of 1946 then retired ゐ (wi) and ゑ (we), whose sounds had merged with い and え, and set the 46 taught today. Japanese children learn all 46 in the first year of primary school, and most know them before they start.

Hiragana vs katakana vs kanji: which script to use when

Japanese writes with three scripts at once, and each has a job. Kanji carry the meaning of nouns and the stems of verbs and adjectives. Hiragana writes everything that holds a sentence together: particles such as は, を, に and の, verb and adjective endings (たべます, たかい), and native words whose kanji is rare or too formal. Katakana writes foreign words and names, sound effects and emphasis. In 私はコーヒーを飲みます 'I drink coffee', the hiragana are は, を and みます.

Hiragana and katakana are twins: the same 46 sounds in the same gojūon order, with the same dakuten and small kana, so あ い う え お matches ア イ ウ エ オ exactly. The shapes differ because they came from different parts of the same Chinese characters: hiragana from whole characters written in cursive, katakana from a single piece of a character written in print style, which is why hiragana curve and katakana are angular. The full katakana set has its own page, Katakana, and both scripts sit together on Japanese Characters.

Hiragana is also the script of readings. Small hiragana printed above a kanji, called furigana, tell children and learners how to read it, and the text in children's books, on the keys of a learner's keyboard and in beginner textbooks is all hiragana. A word written in hiragana instead of its kanji reads as softer, more childlike or more casual, which is why shop names, sweets and mascots often prefer it: おかし 'sweets', ねこ 'cat', ありがとう 'thank you'.

How to type hiragana, and the fonts behind ゐ ゑ and hentaigana

Every phone and computer can type hiragana once the Japanese keyboard is added in the language settings. On a computer you type the sound in Latin letters and the input method turns it into kana as you go: ka gives か, kya gives きゃ, nn gives ん, xtsu or ltu gives a small っ, and xa gives a small ぁ. Pressing Space offers kanji for what you typed; Enter keeps it as hiragana. In Microsoft IME, Alt+` switches between Japanese and English input, and the F6 key converts the current word back to hiragana. On a Mac the Japanese input source switches to hiragana with Ctrl+Shift+J, and on a JIS keyboard with the かな key. Phones offer a 12-key flick keyboard, where you tap あ for a and flick in four directions for い う え お, or the same Latin-letter typing.

The 46 basic kana, their voiced forms and the common small kana have been in Unicode since version 1.0 and are in every system font: Yu Gothic and Meiryo on Windows, Hiragino on Apple devices, Noto Sans CJK on Android and most Linux systems. They always display. Small ゕ ゖ and ゟ were added in Unicode 3.2 (2002) and are safe on anything made this century.

The rest of this page needs newer fonts. The 285 hentaigana and 𛀁 (archaic ye) were encoded in Unicode 10.0 in 2017 at Japan's request, so that pre-1900 texts could be set exactly; the small 𛅐 𛅑 𛅒 came in Unicode 12.0 (2019), and 𛄟 (archaic wu) and small 𛄲 in Unicode 14.0 (2021). Most devices show these as empty boxes unless a font that covers them is installed, such as Noto Serif Hentaigana or BabelStone Han. They copy and paste correctly either way; what you see is a font question, not a character question. One pitfall: Unicode also has combining dakuten and handakuten (U+3099 and U+309A) that attach to the kana before them. They exist for rendering old texts; in ordinary writing always use the ready-made voiced kana such as が and ぱ, which this page gives you, because か followed by a combining mark is a different string that search, sorting and spell-checkers may not recognise.