Japanese Braille (Tenji): Kana Chart, Dakuten and Numbers

Japanese braille, called tenji (点字), writes Japanese by sound, in kana, never in kanji: each basic kana is a single cell. The cell is split into two zones. The vowel sits in dots 1, 2 and 4, and the consonant is added with dots 3, 5 and 6, so か ka is ⠡: the vowel a (dot 1) plus dot 6 for k. Voiced sounds, combined sounds such as kya, and numbers take an extra cell in front. The word tenji itself is ⠟⠴⠐⠳.

We wrote this page in English for teachers, learners of Japanese and braille readers who want to know how another code works. Every cell on it comes from the same table as our Japanese braille translator, a tool that is itself in Japanese.

Key points

• One cell is one kana, that is, one mora, not one letter. Ka takes two cells in English braille and one in Japanese.
• Vowels use dots 1, 2 and 4. Each consonant row adds its own dots from 3, 5 and 6: k adds dot 6, s dots 5-6, t dots 3-5, n dot 3, h dots 3-6, m dots 3-5-6 and r dot 5.
• Dakuten and handakuten become a cell in front of the kana: ⠐ (dot 5) for g, z, d, b and ⠠ (dot 6) for p.
• Yōon such as kya, sho or ju is a prefix, ⠈ (dot 4), plus the a, u or o kana of the row.
• Small っ is ⠂ (dot 2) and the long-vowel mark ー is ⠒ (dots 2-5).
• Numbers use the numeric indicator ⠼ (dots 3-4-5-6) and the cells of a to j, as in English braille, plus a hyphen before some kana.
• There is no kanji in standard Japanese braille. Words are written by their reading, with a space between words.

How the cell works: vowels and consonants

Japanese braille uses the same six-dot cell as English braille, two columns of three: dots 1, 2 and 3 on the left, 4, 5 and 6 on the right. What it adds is a job for each zone. According to the Wikipedia article on Japanese braille, the vowels are written in the upper left, with dots 1, 2 and 4, and can stand alone, while the consonants are written in the lower right, with dots 3, 5 and 6, and never appear on their own.

Here are the five vowels:

The last column matters if you already read English braille: the cells are the same, the meanings are not. Dots 1-2 are b in English and い i in Japanese. Japanese braille was built around the sounds of Japanese, not around the Latin alphabet, so knowing English braille helps with the feel of the dots and very little with the code itself.

The kana chart in Japanese braille

To get any other kana, take the vowel and add the dots of its consonant row. か ka is あ a (dot 1) plus dot 6. き ki is い i (dots 1-2) plus the same dot, which gives dots 1-2-6. Each row keeps its consonant dots across all five vowels:

A few things the chart shows:

• Rows are named by sound, not by romanization. The s-row includes し shi, the t-row includes ち chi and つ tsu, and the h-row includes ふ fu. The cells follow the kana, whatever the spelling in Latin letters.
• The y-row and the w-row are irregular. や ya (dots 3-4), ゆ yu (dots 3-4-6) and よ yo (dots 3-4-5) use dot 4, normally a vowel dot, for the y sound. わ wa is dot 3 and を wo is dots 3-5. These five are simply learned by heart.
• ん n has a cell of its own: ⠴ (dots 3-5-6).
• Katakana uses the same cells as hiragana. Braille does not tell the two scripts apart (Japanese Wikipedia), so カ and か are both ⠡.

That makes 46 basic kana, each written with one cell.

Dakuten and handakuten: a cell in front

In print, the voiced sounds g, z, d and b are marked with two small strokes on the kana (dakuten, ゛), and p with a small circle (handakuten, ゜). In braille, the mark becomes a separate cell placed before the kana: ⠐ (dot 5) for dakuten and ⠠ (dot 6) for handakuten. Wikipedia describes both signs as prefixes. So が ga is the dakuten cell followed by か ka: ⠐⠡.

Every voiced syllable therefore takes two cells, and the kana after the prefix is always the plain one. ぱ pa is the handakuten cell plus は ha: ⠠⠥.

Yōon: kya, sho, ju

Yōon are the combined sounds written in print with a small ゃ, ゅ or ょ after an i-kana: きゃ kya, しょ sho, じゅ ju. Braille writes each one in two cells. First comes the yōon prefix, ⠈ (dot 4), and then the a, u or o kana of the same row. So きゃ kya is the prefix plus か ka: ⠈⠡. The prefix uses the dot that stands for y in や, ゆ and よ, as the Wikipedia article points out.

Voiced and p combinations fold both marks into a single prefix: ⠘ (dots 4-5) for gya, ja and bya, and ⠨ (dots 4-6) for pya.

Notice that the small ゃ, ゅ and ょ never appear as cells of their own: the prefix already says "this is a y combination", and the second cell gives the row and the vowel.

Small tsu, long vowels and spelling by sound

Two more cells come up all the time:

• Small っ (sokuon), which doubles the next consonant, is ⠂ (dot 2). きって kitte, "stamp", is ⠣⠂⠟.
• The long-vowel mark ー (chōon) is ⠒ (dots 2-5). ラーメン rāmen is ⠑⠒⠿⠴.

Braille also follows pronunciation in places where print kana do not. The Japanese Wikipedia article on braille (in Japanese) sets out the rules:

• The particles は and へ are written わ and え, the way they sound.
• Long u and o sounds are written with ー, even where print uses う: 東京 とうきょう becomes とーきょー. Verb endings keep their う, so 思う おもう stays おもう, and words that print spells with おお, such as 大阪 おおさか, keep it too.

A translator program cannot apply these two rules safely on its own, because it would have to know whether a は is a particle or part of a word. Ours leaves them to you: type the braille spelling.

Numbers in Japanese braille

Numbers work as in English braille: the numeric indicator ⠼ (dots 3-4-5-6), then the cells of the letters a to j for 1 to 0. One indicator covers every digit after it, so 2026 is ⠼⠃⠚⠃⠋, the same cells as on our English braille alphabet page.

The difference is what comes right after a number. The a-row (あいうえお) and the ra-row (らりるれろ) use exactly the ten cells of the digits, so a kana from either row right after a number would be read as one more digit. Japanese braille puts a hyphen between them, the tsunagi-fu ⠤ (dots 3-6). Both Wikipedia, with the example of 6円, and the Japanese Wikipedia give the rule.

Latin letters inside Japanese text work in a similar way. They take the foreign-letter sign, gaiji-fu, ⠰ (dots 5-6), followed by the letters, which use the same cells as in English.

Why Japanese braille has no kanji

Printed Japanese mixes kanji with two kana scripts. Braille keeps only the sound, for three reasons:

1. The cell is too small. Six dots give 63 patterns with at least one dot raised. The list of kanji for everyday use alone, the jōyō kanji, has 2,136 characters (Wikipedia).
2. Every word can be written by its reading. Any kanji word has a pronunciation, and the pronunciation fits in kana. 点字 is written てんじ, and 日本 is written にほん or にっぽん.
3. Spaces do part of the job. Braille has no kanji and does not distinguish hiragana from katakana, so it leaves one blank cell between word units to make text easier to follow, a practice called wakachigaki (Japanese Wikipedia).

The cost is homophones. はし hashi can be 橋 "bridge", 箸 "chopsticks" or 端 "edge" (Wiktionary), and in braille all three are ⠥⠳. Readers tell them apart from context, as listeners do in conversation.

Kanji braille does exist, but it is not the standard. Kantenji, devised in 1969 by Tai'ichi Kawakami, a teacher at the Osaka School for the Blind, uses an eight-dot cell and writes each kanji with one to three cells. Wikipedia lists its status as unofficial, and the article on Japanese braille calls the six-dot kana system the significantly more common one.

Where Japanese braille comes from

Japanese braille is an adaptation of the six-dot cell to kana, made at the Tokyo School for the Blind and Deaf. A committee there considered at least four proposals and, in 1890, chose the one by teacher Ishikawa Kuraji (1859–1944), as Wei Yu Wayne Tan recounts in History Workshop. An expanded version was approved in 1898. The choice was made on November 1, which is now Japan's braille day, according to the city of Hamamatsu, where Ishikawa was from. The system was published in Japan's official gazette in April 1901 as Nihon kunmō tenji (Japanese Wikipedia).

For how Japanese compares with Korean, Chinese, Arabic and the Latin-based codes, see braille in other languages.

Japanese braille and English braille side by side

Try it and learn it

• Type kana into our Japanese braille translator, which is in Japanese. It converts hiragana, katakana, numbers and Latin letters, and leaves kanji out, so type the reading. Apply the two spelling rules above yourself.
• To compare with English, type a word into the English braille translator or look up the vowel cells above in the English braille alphabet.
• To learn it in an app: with the app in Japanese, Braillito teaches Japanese braille in 76 lessons across 9 worlds. The lessons are exercises you play, with lives, levels and a daily streak, and the app also has a translator, a dictionary and a builder. It is a subscription with no free version: depending on the plan, first-time subscribers get a free trial or a cheaper first period.

Frequently asked questions

Is Japanese braille the same as English braille?

No. Both use the same six-dot cell, but Japanese braille assigns the cells to kana. Dots 1-2, b in English, are い i in Japanese. Only numbers and Latin letters inside Japanese text keep the meanings an English braille reader knows.

How many characters does the Japanese braille alphabet have?

Strictly speaking it is a syllabary, not an alphabet. The basic chart has 46 kana, one cell each. Voiced sounds, p sounds and yōon add a prefix cell to those, and small っ and the long-vowel mark have a cell each.

Does Japanese braille have kanji?

Standard Japanese braille does not: every word is written by its reading, in kana, with spaces between words. An eight-dot kanji system, kantenji, exists, but it is unofficial and much less common.

How do you write "konnichiwa" in Japanese braille?

As こんにちわ, with わ instead of は, because braille follows pronunciation: ⠪⠴⠇⠗⠄.

How do you write "Japan" in Japanese braille?

日本 is written by its reading, にほん ⠇⠮⠴ or にっぽん ⠇⠂⠠⠮⠴.

Who invented Japanese braille?

Ishikawa Kuraji, a teacher at the Tokyo School for the Blind and Deaf, whose proposal was chosen in 1890.