Is Braille the Same in Every Language?

Braille is not the same in every language. What is universal is the medium: the six-dot cell, in two columns of three, with its 64 combinations. What changes from one language to another is what each combination means. Someone who reads English braille can run a finger over a page of Japanese braille and feel every dot perfectly, and understand nothing: the same bumps, with a different code underneath.

It is an understandable confusion, because "braille" is talked about as if it were a language, and it is not. Braille is a writing medium, like the Latin alphabet. The Latin alphabet writes English, Vietnamese and Turkish, and reading English does not let you read Turkish. Braille works exactly the same way, with one extra twist: in some languages the cell does not even stand for a letter.

Key points

• The cell is universal; the code is not. Six dots, two columns of three, 64 combinations including the empty cell. That is the same everywhere.
• Dot numbering is the same everywhere too: 1, 2, 3 down the left column; 4, 5, 6 down the right.
• All braille is read left to right, including Arabic and Hebrew, whose print runs right to left.
• What changes is the assignment: which letter, syllable or sound each cell stands for, and whether the system encodes letters, syllables or sounds.
• French is the common ancestor of Latin-alphabet braille: a congress in Paris in 1878 adopted the French values as the standard.
• Japanese, Korean and Chinese braille do not follow that framework. They are independent systems built on the sounds of each language.
• In Japanese, the vowel takes dots 1-2-4 and the consonant dots 3-5-6, so a whole syllable fits in one cell.
• In Korean, each syllable is broken into jamo, and the same consonant has different cells at the start and at the end of a syllable.
• The standard reference, World Braille Usage (3rd edition, 2013), documents 133 languages in 137 braille codes across 142 countries.
• Unicode confirms it from the other side: it encodes the dot pattern, not the letter. U+2817 ⠗ is "braille pattern dots-1235", not "the letter R".

What is the same in every language

There is one layer that does not change in any braille system in the world, and it is worth fixing it before talking about differences.

Dot numbering deserves its own sentence, because it is where popular explanations most often go wrong. Dots are always numbered the same way: the left column is 1 at the top, 2 in the middle, 3 at the bottom; the right column is 4, 5 and 6. No language reverses that numbering or rotates the cell. That consistency is what lets "dots 1-2-4" mean the same thing in any country.

The cleanest proof that the cell is universal and the meaning is not comes from Unicode. The Braille Patterns block (U+2800–U+28FF) contains all 256 possible eight-dot patterns, and the standard is explicit: "In Unicode, a braille cell does not have a letter or meaning defined. For example, Unicode does not define U+2817 ⠗ BRAILLE PATTERN DOTS-1235 to be "R"" (Braille Patterns). Unicode encodes holes in the paper. The language supplies the meaning.

One cell, six meanings

The quickest way to see the problem is to take one cell and look at what it means in each place. Take dots 1-2-4-5-6 (⠻).

Source: the values of the braille pattern dots-12456. The dots are the same; the finger feels exactly the same thing. Underneath are six different conventions, and one of them, English, is not even a letter but a group of two.

What really changes from one language to another

Four things are usually lumped together, and they are worth separating.

1. The cell assignments. The most obvious: which cell is "a", which is "b". Latin alphabets share a lot through their French inheritance; outside them, almost nothing.

2. What unit a cell stands for. This is the big difference, and the one almost nobody explains. A cell can stand for:

• a letter (English, Spanish, French, German, Russian…),
• a syllable or mora (Japanese),
• a part of a syllable (Korean: initial consonant, vowel, final consonant; Chinese: initial, final, tone),
• or a whole group of letters, in contracted systems.

3. Indicators. Capital, number, italics, quotation marks, opening question marks. These vary more than people expect, even between neighbouring languages.

4. Contractions. English has a contraction system in general use; Spanish has an approved one that is barely used. We cover English contractions in braille contractions.

Why French is the common ancestor

The braille of almost every Latin-alphabet language descends from French braille because an international congress in Paris decided so in 1878. Before that, incompatible national conventions coexisted.

The Wikipedia article on the international uniformity of braille alphabets documents the decision: representatives of France, the United Kingdom, Germany and Egypt agreed to replace those divergent conventions with the French values of the basic Latin alphabet, adopting the system "with the values of its symbols unaltered from those of the original French". The American Foundation for the Blind confirms the date: "In 1878, a congress met in Paris and officially decided to adopt braille as the international system used for writing by the blind".

That is where the structure anyone who has studied the alphabet recognises comes from: the first ten letters use only the top four dots, and the following series repeat that pattern with dots added below. It is the order Louis Braille designed for French, and the rest inherited it.

The agreement had two important limits, and both still apply:

• It did not cover letters beyond the basic 26. Each language solved its own accented and extra letters, and that is where compatibility breaks.
• It did not solve non-Latin alphabets. For those, the criterion was to assign cells by transliteration: the Greek gamma γ gets the cell of "g" because of its romanisation, not its position in the alphabet. At the UNESCO congress of 1951 it was recognised that most non-European alphabets had letters with no braille precedent, which had to be assigned more or less arbitrarily.

That transliteration criterion explains why Russian, Greek, Hebrew and Arabic braille are still recognisable to someone who knows Latin braille, and why Japanese, Korean and Chinese are not at all.

Spanish: how French braille became Spanish

Spanish braille was built on French braille, and ñ takes the cell that was ï in French. It is not a curiosity: it is the visible trace of the family tie.

The Instituto Cervantes 2023 Yearbook, in Andrés Ramos Vázquez's study of braille (in Spanish), documents it precisely: the braille alphabet for the languages of Spain is based, as in most languages with a Latin alphabet, on the original French braille alphabet. And it details the adaptation:

According to the same study, of the nine original series, Spanish kept seven, and the first four, 40 characters, match the French braille alphabet completely.

Forty identical characters. A Spanish reader and a French reader share almost the whole basic alphabet and part ways at the accents. That is the real distance between two related systems. The full Spanish alphabet, with ñ and the accented vowels, is in the Spanish braille alphabet.

English: the same alphabet, a different problem

English shares the basic alphabet with Spanish and French through the same French inheritance, so the letters transfer almost entirely. What does not transfer is the rest of the code.

The current standard is Unified English Braille (UEB), and its distinctive feature is not the letters but the contractions: in English, contracted braille is the publishing standard, to the point that learning English braille without contractions leaves you out of almost every library. A Spanish braille reader who opens a book in English braille will recognise the letters and not understand the text, because many of the cells are not letters but words and letter groups. See braille contractions, and try any sentence in our braille translator.

Japanese: the cell is a syllable, not a letter

Japanese braille, called tenji (点字), does not encode letters: it encodes the syllables of kana. And its internal design has nothing to do with Latin braille.

It was proposed by Ishikawa Kuraji (1859–1944), a teacher at the school for the blind and deaf in Tokyo, and chosen from at least four proposals at the final meeting of 1890, according to History Workshop, which places the work in collaboration with Konishi Nobuhachi.

The idea is elegant: an open Japanese syllable is consonant + vowel, so the cell is split into two zones.

• The vowel takes the upper left: dots 1, 2 and 4. It can stand alone.
• The consonant takes the lower right: dots 3, 5 and 6. It cannot stand alone.

As the Wikipedia article on Japanese braille puts it: "The vowels are written in the upper left corner (dots 1, 2, 4) and may be used alone. The consonants are written in the lower right corner (dots 3, 5, 6) and cannot occur alone".

The five vowels look like this, and this is where you see how different the code is:

A whole mora in one cell

To write a syllable, tenji does not chain two cells: it adds the vowel dots and the consonant dots inside the same cell. The ka series adds dot 6 to each vowel:

Each consonant series has its own mark in the lower half: k adds dot 6, s adds 5-6, t adds 3-5, n adds 3, h adds 3-6, m adds 3-5-6 and r adds 5. So a whole Japanese mora fits in one cell, while in English the same syllable would take two.

Three more details from the same source: voiced sounds (dakuon), half-voiced sounds (handakuon) and yōon are marked with a prefix cell before the syllable; the long vowel is written with the sign ⠒ (chōon); and tenji separates words with spaces, which printed Japanese does not.

Korean: the cell breaks hangul into jamo

Korean braille breaks each hangul syllable into its jamo (initial consonant, vowel, final consonant) and gives each one a cell. As a result, most syllables take two or three cells.

It was created by Park Du-seong, a teacher of blind pupils, and presented on November 4, 1926, under the name Hunmaengjeongeum (훈맹정음), according to Korea.net, the Republic of Korea's official portal. The Wikipedia article on Korean braille documents the structure and dates a revision of the current form to 1994.

The current standard is the Korean Braille Regulation (한국 점자 규정), published by the National Institute of the Korean Language and approved by the Ministry of Culture, Sports and Tourism in notice 2020-38, in force since September 16, 2020. It covers hangul as well as mathematical, scientific, computer and music notation.

The detail that makes the system interesting: the same consonant gets different cells depending on whether it starts or ends a syllable. The shapes are the same, shifted within the cell: "the initial and final variants have the same shapes, but are shifted across the braille block". It uses the geometry of the six dots to encode position, something no Latin braille does.

It helps to see why breaking syllables apart was necessary. Hangul does not have a closed set of letters to tabulate: it combines 19 initial consonants × 21 vowels × 28 finals (counting no final), which gives 11,172 possible syllables. No table of 64 cells can cover that. Breaking into jamo is not elegance: it is the only way out. Korean braille also has abbreviations for frequent combinations and common grammatical words.

Arabic: where almost everyone gets it wrong

Arabic braille is read from left to right, like every other braille. This is the claim in this article that most contradicts what circulates online, and it is confirmed by two independent sources.

The Unified Arabic Braille portal of Qatar's Mada Center says it plainly: "braille in Arabic is read from left to right and not the opposite as it should be in Arabic text, this is to be in line with the braille reading format of all languages". The Wikipedia article on Arabic braille agrees: "Unlike the Arabic script, Arabic Braille is read from left to right, following the international convention".

And it is not an Arabic exception. Hebrew braille does the same: it is read left to right even though Hebrew print runs right to left (Hebrew Braille).

The underlying reason is tactile, not political. The reading finger sweeps the line in one direction, and changing that direction by language would have meant redesigning writing, printing and teaching for no gain.

Three things that are true of Arabic braille:

• It is not fully unified. A conference in Saudi Arabia in 2002 set a unified standard, but by 2013 not every country had adopted it, including some Arab countries and non-Arab Muslim countries such as Iran, Malaysia and Indonesia (Wikipedia, Arabic Braille).
• Short vowels (harakat) are usually left out, as in Arabic print. When they are written, they are not marks attached to the letter: they take their own cell, placed after the consonant.
• The shadda, which marks a doubled consonant, is the exception: it is written before the letter it affects.

That last point illustrates something general. In braille, a feature that sits above or below a letter in print almost always becomes a separate cell on the line, because in braille there is no "above" or "below": only before and after.

Chinese: the most complex case, with more than one system

In Chinese there is not one braille but several, and none of them encodes characters. They all encode the sound of the syllable, because the characters number in the thousands and Mandarin syllables are around thirteen hundred.

The main system in mainland China writes each syllable with up to three cells: initial, final and tone, according to the Wikipedia article on mainland Chinese braille, which also notes that the tone is usually left out, as in everyday pinyin. So in practice two cells per syllable is normal, and the third appears only when needed to resolve ambiguity.

There are also, as separate systems:

• Two-cell Chinese braille, designed in the 1970s, where the first cell combines initial and medial and the second combines rhyme and tone (Two-cell Chinese Braille).
• Taiwanese braille, used for Mandarin in Taiwan.
• Cantonese braille.
• The National Common Braille (国家通用盲文), published in July 2018 by the Ministry of Education, the State Language Commission and the China Disabled Persons' Federation; its official document is published by the Chinese Ministry of Education. Unlike the earlier system, it marks the tone on every syllable.

How many braille systems are there?

We do not give a single figure for "how many brailles exist", because it depends on what you count. What can be cited is the standard reference.

World Braille Usage, in its third edition of 2013, published by Perkins with the International Council on English Braille and the Library of Congress's National Library Service for the Blind and Physically Handicapped, with rights from UNESCO, documents 133 languages transcribed in 137 alphabet and punctuation braille codes, representing 142 countries (Paths to Literacy). The second edition, of 1990, covered 97 languages; the first is from 1953 (ABILITY Magazine).

That there are 137 codes and not 133 already says something: some languages have more than one braille code in use, as we just saw with Chinese.

What transfers if you already know braille

This is the practical question that brings most people to this article. The short answer: more than it seems within Latin alphabets, and almost nothing outside them.

The numeric indicator shows the nuance well. The cell of dots 3-4-5-6 (⠼) works as the numeric indicator in French, English, German, Russian, Persian and Japanese braille, with digits then written using the first ten letters. But it is not universal: in mainland Chinese braille the same cell is the final "eng", and in Luxembourgish braille it is the digit 0 (braille pattern dots-3456).

In other words: if you learn braille in one language, you have done the most costly part, reading with your fingers, and much of the Latin alphabet. What each new language asks of you is memorising assignments, and in Japanese, Korean or Chinese, also understanding what unit the cell stands for.

If you are starting out, the sensible order is the usual one: one language in depth first, with how to learn braille as an adult, and the others later.

What we could not confirm

This article covers systems that are rarely documented in English outside specialist sources, and secondary accounts repeat second-hand mistakes. These are the seams.

Frequently asked questions

Is braille the same in every language?

No. The six-dot cell is the same all over the world, and so are the dot numbering and the reading direction, left to right. What changes is what each combination means, and in some languages even what kind of unit the cell stands for: a letter in English, a syllable in Japanese, a part of a syllable in Korean and Chinese.

Can an English braille reader read braille in another language?

It depends on the language. In another Latin-alphabet language they will recognise almost all the basic letters, which all come from French braille through the 1878 agreement, and stumble on accents, punctuation and contractions. In Japanese, Korean or Chinese they will recognise nothing: those systems are built on other sound systems, outside the international alphabetic framework.

Is Arabic braille read right to left?

No. Arabic braille is read and written left to right, following the international convention, even though Arabic print runs right to left. The dot numbering is not reversed and the cell is not mirrored. Hebrew braille works the same way.

How does Japanese braille work?

Japanese braille, or tenji, encodes kana syllables instead of letters. The cell is split into two zones: the vowel takes dots 1, 2 and 4 and can stand alone; the consonant takes dots 3, 5 and 6 and never stands alone. The two are combined in the same cell, so a whole syllable takes a single cell: か ka is dots 1-6, which is あ a (dot 1) plus the mark of the k series (dot 6). Ishikawa Kuraji proposed it and it was adopted in 1890.

And Chinese braille?

Chinese braille does not represent characters but sounds. The main system in mainland China writes each syllable with up to three cells, initial, final and tone, and the tone is usually left out. There are also two-cell Chinese braille, Taiwanese braille, Cantonese braille and the National Common Braille published in 2018, which marks the tone on every syllable.

Is there a universal braille for all languages?

No, and there never has been, despite several attempts: the Paris congress of 1878, the 1929 conference and the UNESCO congresses of 1950 and 1951. What was achieved was unifying the values of the basic Latin alphabet and a transliteration criterion for non-Latin alphabets. Languages whose writing is not alphabetic stayed outside that framework by necessity.

Learning braille in the language you need

If you have read this far, the practical conclusion is simple: there is no learning "braille". You learn the braille of a language, and then add others with much less effort than the first one took.

That is why Braillito teaches 38 languages with their real code, not the English or Spanish alphabet in disguise. It is a distinction that sounds small and is not: teaching "Japanese braille" with Latin cells produces someone who cannot read a sign in Tokyo. Non-Latin braille systems are not cultural curiosities; they are the everyday writing of millions of people, and they deserve to be taught as what they are.