From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

The sokuon (促音) is a Japanese symbol in the form of a small hiragana or katakana tsu. In less formal language it is called chiisai tsu (小さいつ) or chiisana tsu (小さなつ), meaning "small tsu".[1] It serves multiple purposes in Japanese writing.


In both hiragana and katakana, the sokuon appears as a tsu reduced in size:

Full-sized Sokuon

Use in Japanese[edit]

The main use of the sokuon is to mark a geminate consonant,[1] which is represented in most romanization systems by the doubling of the consonant, except that Hepburn romanization writes a geminate ch as tch. It denotes the gemination of the initial consonant of the symbol that follows it.


  • Pocky, a Japanese snack food, is written in kana as ポッキー, which is
    In rōmaji, this is written pokkī, with the sokuon represented by the doubled k consonant.
  • 待って (matte), the te form of the verb 待つ (matsu, "wait"), is composed of:
    ma (kanji)
    In the rōmaji rendering, matte, the sokuon is represented by the doubling of the t consonant.
  • こっち (kotchi), meaning "here", is composed of:
    In Hepburn romanization, kotchi, the sokuon is represented by the t consonant, even though the following consonant is ch. This is because rōmaji ch actually represents [t͡ɕ] (voiceless alveolo-palatal affricate), and the sokuon before it doubles the [t] sound. The Kunrei-shiki and Nihon-shiki romanization systems write this syllable as ti (and its geminate version as tti) so the exception does not arise.

The sokuon never appears at the beginning of a word or before a vowel (a, i, u, e, or o), and rarely appears before a syllable that begins with the consonants n, m, r, w, or y. (In words and loanwords that require geminating these consonants, n, mu, ru, u, and i are usually used, respectively, instead of the sokuon.) In addition, it does not appear before voiced consonants (g, z, d, or b), or before h, except in loanwords, or distorted speech, or dialects. However, uncommon exceptions exist for stylistic reasons: For example, the Japanese name of the Pokémon species Cramorant is ウッウ, pronounced /uʔu/.[2]

The sokuon is also used at the end of a sentence, to indicate a glottal stop (IPA [ʔ], a sharp or cut-off articulation),[3] which may indicate angry or surprised speech. This pronunciation is also used for exceptions mentioned before (e.g. a sokuon before a vowel kana). There is no standard way of romanizing the sokuon that is at the end of a sentence. In English writing,[clarification needed] this is often rendered as an em dash. Other conventions are to render it as t or as an apostrophe.

In the International Phonetic Alphabet, the sokuon is transcribed with either a colon-like length mark or a doubled consonant:

  • kite (来て, "come") – /kite/
  • kitte (切手, "postage stamp") – /kitːe/ or /kitte/
  • asari (あさり, "clams") – /asari/
  • assari (あっさり, "easily") – /asːari/ or /assari/

The sokuon represents a mora, thus for example the word Nippon (Japan) consists of only two syllables, but four morae: ni-p-po-n.[4]

Use in other languages[edit]

In addition to Japanese, sokuon is used in Okinawan katakana orthographies. Ainu katakana uses a small ッ both for a final t-sound and to represent a sokuon (there is no ambiguity however, as gemination is allophonic with syllable-final t).

Computer input[edit]

There are several methods of entering the sokuon using a computer or word-processor, such as xtu, ltu, ltsu, etc. Some systems, such as Kotoeri for macOS and the Microsoft IME, generate a sokuon if an applicable consonant letter is typed twice; for example tta generates った.

Other representations[edit]


⠂ (braille pattern dots-2)

Character information
Encodings decimal hex dec hex dec hex
Unicode 12387 U+3063 12483 U+30C3 65391 U+FF6F
UTF-8 227 129 163 E3 81 A3 227 131 131 E3 83 83 239 189 175 EF BD AF
GB 18030 164 195 A4 C3 165 195 A5 C3 132 49 151 49 84 31 97 31
Numeric character reference っ っ ッ ッ ッ ッ
Shift JIS[5] 130 193 82 C1 131 98 83 62 175 AF
EUC-JP[6] 164 195 A4 C3 165 195 A5 C3 142 175 8E AF
EUC-KR[7] / UHC[8] 170 195 AA C3 171 195 AB C3
Big5 (non-ETEN kana)[9] 198 199 C6 C7 199 91 C7 5B
Big5 (ETEN / HKSCS)[10] 199 74 C7 4A 199 191 C7 BF


  1. ^ a b Kawahara, Shigeto. "The phonetics of obstruent geminates, sokuon" (PDF). Semantic Scholar. S2CID 145942. Archived from the original (PDF) on 2019-07-18. Retrieved 2019-06-28. {{cite journal}}: Cite journal requires |journal= (help)
  2. ^ The pronunciation is verifiable here: Nintendo Direct (September 5, 2019; 23 min 48 s). Retrieved 2019-09-05.
  3. ^ "What is that small tsu at the end of a sentence?". Retrieved 2019-06-28.
  4. ^ Nick Miller, Anja Lowit (2014). Motor Speech Disorders: A Cross-Language Perspective. Multilingual Matters. p. 223. ISBN 9781783092321.
  5. ^ Unicode Consortium (2015-12-02) [1994-03-08]. "Shift-JIS to Unicode".
  6. ^ Unicode Consortium; IBM. "EUC-JP-2007". International Components for Unicode.
  7. ^ Unicode Consortium; IBM. "IBM-970". International Components for Unicode.
  8. ^ Steele, Shawn (2000). "cp949 to Unicode table". Microsoft / Unicode Consortium.
  9. ^ Unicode Consortium (2015-12-02) [1994-02-11]. "BIG5 to Unicode table (complete)".
  10. ^ van Kesteren, Anne. "big5". Encoding Standard. WHATWG.
  • Fujihiko Kaneda, Rika Samidori (1989). Easy hiragana: first steps to reading and writing basic Japanese. Passport Books. pp. 74−78.

See also[edit]

External links[edit]