Morse code beyond the Latin alphabet: Cyrillic, Arabic, Chinese and more
Morse code was built around the 26 letters of the English alphabet. So what happens when the message is not in English — when it is Russian, Greek, Arabic or Chinese? The answer is a short tour of the tricks people have used to fit the world’s writing into dots and dashes.
The easy case: alphabets near Latin
For a language written with a few dozen letters, the fix is simple — give each letter its own dot-dash pattern, usually borrowing the pattern of a similar-sounding Latin letter. Russian Morse does this for Cyrillic: А is , the same as Latin A; Б (a “b” sound) takes , the pattern of Latin B; and so on, with extra patterns for the letters Cyrillic has and Latin does not. Greek has its own long-standing table. Hebrew and Arabic each have a full Morse mapping, with a code for every letter of the script. The message still goes out one letter at a time; only the letter-to-pattern table changes.
The hard case: thousands of characters
Chinese breaks the whole approach. You cannot hand a unique dot-dash pattern to tens of thousands of characters — no one could learn it, and the long ones would take forever to key. The solution, worked out around 1871 by a Frenchman named Septime Viguier building on earlier efforts, was to put a codebook in the middle. Every common character is given a four-digit number; to send it, you look up the number and transmit the four digits in ordinary Morse, and the operator at the far end reads the digits and looks the character back up. It is slow, it needs a book at both ends, and it is a lovely early example of something computers now do constantly — encoding a huge set of characters as plain numbers.
Japanese took a different road. Instead of numbers, Wabun code assigns dot-dash patterns to the kana — the syllables of Japanese — rather than to Latin letters. A Wabun operator sending Japanese and a Western operator sending English are both “doing Morse,” but the same pattern means completely different things, so you have to know which code is in play before you can read a word. Korean uses similar syllable-based schemes.
What this says about Morse itself
Morse handles a few dozen symbols comfortably — letters, digits and a handful of punctuation marks, which is exactly what you will find on the reference table. Push past that and you need one of two tricks: remap the patterns onto a new alphabet, or add a codebook that turns characters into numbers first. Those are still the two basic ways any system copes with more symbols than it has room for.
A note on the app
DotDotDoit teaches International Morse — the Latin letters, the digits and the shared punctuation. That is the code in the trainer and on the reference table. The app’s menus are translated into 14 languages so it is comfortable to use wherever you are; the Morse it sends and reads is the international standard. The Cyrillic and Wabun tables are a rewarding rabbit hole if you are curious — but the international code is the one that lets a ham in Japan and a ham in Brazil work each other at two in the morning.
Keep going: browse the interactive alphabet, play with the translator, or get the app to practise properly.
More guides: The history of Morse code · Morse code vs the telegraph.