Japanese Morse Code: Wabun Chart and Translator

The translator above is open on the Japanese (Wabun) table for this visit only, so your usual table stays as you left it. What it accepts:

• Hiragana and katakana. こんにちは and コンニチハ give the same code, because hiragana is read as the matching katakana.
• Voiced kana. が, ぱ and the other voiced and p-sound kana are split into the base kana plus ゛ or ゜, the way Wabun sends them.
• Small kana. っ, ゃ, ゅ, ょ and the other small kana are sent as their full-size forms.
• Marks and numbers. ー, the Japanese comma 、 and full stop 。, and the digits 0 to 9.

Kanji and romaji have no Wabun codes. Typed in, 東京 gets the message "No Morse code in the Japanese (Wabun) table for 東 京, so they’re left out." Type the reading instead: とうきょう.

Switch the direction to read Wabun back. The decoder joins the voicing mark to the kana before it, so -.-.. .. --.-. -.-.. reads ギシキ, not キ゛シキ.

Wabun also has codes for opening and closing brackets, and operators use two signals, DO and SN, to switch between codes (see below). The translator's Wabun table doesn't include these: pasted in as Morse in Wabun mode, they show as ?.

The whole code: 48 kana in iroha order (イ, ロ, ハ and on), the order the code was built on in 1873, then the voicing marks, the long-vowel mark and the two punctuation marks. Each row plays.

• イi
  .-
• ロro
  .-.-
• ハha
  -...
• ニni
  -.-.
• ホho
  -..
• ヘhe
  .
• トto
  ..-..
• チchi
  ..-.
• リri
  --.
• ヌnu
  ....
• ルru
  -.--.
• ヲwo
  .---
• ワwa
  -.-
• カka
  .-..
• ヨyo
  --
• タta
  -.
• レre
  ---
• ソso
  ---.
• ツtsu
  .--.
• ネne
  --.-
• ナna
  .-.
• ラra
  ...
• ムmu
  -
• ウu
  ..-
• ヰwi
  .-..-
• ノno
  ..--
• オo
  .-...
• クku
  ...-
• ヤya
  .--
• マma
  -..-
• ケke
  -.--
• フfu
  --..
• コko
  ----
• エe
  -.---
• テte
  .-.--
• アa
  --.--
• サsa
  -.-.-
• キki
  -.-..
• ユyu
  -..--
• メme
  -...-
• ミmi
  ..-.-
• シshi
  --.-.
• ヱwe
  .--..
• ヒhi
  --..-
• モmo
  -..-.
• セse
  .---.
• スsu
  ---.-
• ンn
  .-.-.
• ゛voicing mark
  ..
• ゜p-sound mark
  ..--.
• ーlong vowel
  .--.-
• 、comma
  .-.-.-
• 。full stop
  .-.-..

The numbers 0 to 9 use the same codes as International Morse, so 1 is .---- in both.

Wabun has a code for every basic kana, not for every sound. Three rules cover the rest.

Voiced sounds take a second code. ガ is sent as カ followed by the voicing mark ゛ (..), and パ as ハ followed by ゜ (..--.). Each mark is its own character, with a normal letter gap before it.

• ガ (カ + ゛)

  .-.. ..15 units · 0.90 s

• パ (ハ + ゜)

  -... ..--.25 units · 1.50 s

Small kana are sent full size. The Japanese table defines no codes for small ゃ, ゅ, ょ or っ. They go as ヤ, ユ, ヨ and ツ. Japanese Wikipedia's example is ヒャッカジテン (encyclopedia), sent as ヒヤツカシ゛テン. The site's encoder gives exactly that:

• キャ (キ + ヤ)

  -.-.. .--25 units · 1.50 s

• ヒャッカジテン

  --..- .-- .--. .-.. --.-. .. .-.-- .-.-.111 units · 6.66 s

Kanji are sent by their reading. Kana has only 48 characters; kanji number in the thousands. So a word written in kanji goes out as its kana spelling: 東京 as とうきょう. The long-vowel mark ー, common in katakana loanwords such as コーヒー, has a code of its own (.--.-).

Real output from the site's encoder, with the length in units (one unit is a dot) and the time at 20 words per minute. Press play on any code.

Two things stand out. The voicing mark adds a whole character, so ありがとう is five kana but six codes. And the long-vowel mark counts as a character too: コーヒー has four, and takes as long to send as こんにちは.

The site sends Wabun with the timing ITU-R M.1677-1 sets for International Morse: a dash is three dots long, with gaps of one dot inside a character, three between characters and seven between words. What changes is the meaning. Most Wabun codes are also International codes, so the same signal reads differently depending on which table the operator has in mind:

Japanese Wikipedia notes that the codes take no account of how often each kana is used: ヘ has the single dot, while コ takes four dashes (----).

Operators mixing the two codes need a way to switch. The prosign DO announces the start of Wabun, and SN announces the return to International Morse. In the translator's International table you can type them as <DO> and <SN>. In Wabun mode they are not read as switch signals.

• DO (switch to Wabun)

  -..---19 units · 1.14 s

• SN (back to International)

  ...-.11 units · 0.66 s

How International Morse works on its own is covered in What is Morse code?, and the full Latin chart is on the Morse code alphabet page. Wabun is not the only national code: the US railroads and telegraph lines used American Morse code, with different letters and spacing.

Morse reached Japan as a gift. Commodore Perry's expedition presented a Morse telegraph to the government in 1854. A Dutch telegraph followed with the first Morse code for katakana, and in 1855 Japanese trainees demonstrated it for the shogun, Tokugawa Iesada. A second Japanese code followed in 1869.

The code in use today is based on one devised in 1873 by Yoshida Masahide, Terasaki Yuzuru and others. They took the April 1854 code of the German-Austrian Telegraph Union and assigned the kana to it in iroha order, イ to the first code, ロ to the next, and so on. Later changes:

Today the code is set out in Table 1 of Japan's Radio Station Operation Regulations (無線局運用規則 別表第1号), next to the International code.

In September 1942, the American radio magazine QST explained the code to its readers: Japanese Morse was used on Japan's landline telegraph and by Japanese ship and fixed radio stations, while messages to foreign countries or ships went in International Morse. How the armed forces of other countries used Morse is on Military Morse code. Whether a code like this counts as a language of its own is taken up in Is Morse code a language?

Type your name in kana into the translator at the top of this page and press Play, or switch it to Morse to text and decode a message from a friend. The full Morse code translator has the Japanese (Wabun) table too, under Code table, along with sound, light and vibration and an MP3 download.