Katakana

• 1. Introduction

  ›

  Katakana is a Japanese syllabary that carries a specific, practical purpose: marking the foreign, the technical, and the unusual within the Japanese writing system. Its very name tells you what it is. Katakana means "fragmentary kana," because its characters were drawn from pieces of older, more complex Chinese characters. Where hiragana handles native Japanese words and grammatical particles, katakana signals something different: a borrowed word, a species name, a sound effect, a company logo. The system sits at the intersection of phonetics and culture, doing work that English would accomplish with italics. How did Buddhist monks in 9th-century Nara give rise to the script that now spells out "ice cream" and "Toyota" on Japanese streets? And why does a system devised a thousand years ago still live inside Unicode, on convenience store receipts, and in the subtitles of Japanese digital television? Those are the questions worth following.

• 2. Structure of the Script

  ›

  The complete katakana script consists of 48 characters. Those characters break down into five nucleus vowels, 42 core syllabograms built from nine consonants paired with each of the five vowels, and one coda consonant. Three theoretically possible combinations, yi, ye, and wu, were never standardized and do not appear in modern Japanese. The 48 sounds are organized into a 5-by-10 grid called the gojūon, which literally means "fifty sounds." The vowel and consonant order in that grid was inherited from Sanskrit practice, not from any native Japanese tradition. Reading across from the top, the grid runs ア (a), イ (i), ウ (u), エ (e), オ (o), カ (ka), キ (ki), and onward through nine consonant rows. Two diacritic marks sit at the upper right of any base character to shift its initial sound. A double dot called the dakuten voices the consonant: ka becomes ga, sa becomes za, ta becomes da, ha becomes ba. A circular mark called the handakuten applies only to the h-row, pushing h to p, so ha becomes pa. These diacritics have existed for over a thousand years, but only became mandatory in the Japanese writing system in the second half of the 20th century. When a small version of tsu, called the sokuon, appears between two characters, it doubles the consonant that follows. The word saka, meaning hill, becomes sakka, meaning author, with the addition of that single small character.

• 3. The Uses of Katakana

  ›

  "Ice cream" in Japanese is written in katakana. So is "Homo sapiens." So is the "ding-dong" of a doorbell. The range of what katakana is expected to carry in modern writing is wide and deliberate. Foreign loanwords, called gairaigo, are its primary domain: country names, foreign cities, and personal names from outside Japan all appear in katakana. The company name Suzuki is written スズキ in katakana rather than kanji, which matters because Suzuki is also the second most common family name in Japan. Writing the company name in katakana separates it visually from the surname. Toyota is written トヨタ for the same reason. Technical and scientific terms follow the same convention; animal and plant species names are commonly given in katakana. Before 1988, telegrams in Japan were transmitted in katakana. Through the 1980s, most computers used katakana for output rather than the more complex kanji or hiragana, because the technology of the day handled it more easily. Pre-World War II official documents of the Empire of Japan mixed katakana with kanji the same way modern Japanese mixes hiragana with kanji, using katakana for grammatical elements rather than a separate writing mode. In manga, a foreign character or a robot might speak in katakana to flag their accent or non-human nature; what appears in katakana signals to the reader that this is a translation of something not originally Japanese. Medical terminology offers another window into the system's reach: the kanji for the second character in the word for dermatology is considered difficult to read, so the word is commonly written mixing kanji with katakana, swapping the hard character out.

• 4. Borrowed Sounds, Extended Characters

  ›

  Japanese phonology does not naturally include a sound like the English "che" in "change," or the "fa" in "family." Katakana solves this by combining characters in ways the standard grid never specified. Small versions of the five vowel kana are appended to existing characters to create digraphs: チェ (che) appears in チェンジ, the katakana rendering of "change," and ファ (fa) opens ファミリー, the katakana for "family." The word Wikipedia itself is rendered ウィキペディア using these extended combinations. A short line called the chōonpu, or "long vowel mark," handles stretched vowels in foreign words. In horizontal text the line runs horizontally; in vertical text it runs vertically. The gairaigo for email, メール, uses the chōonpu to hold the long e sound from the English word "mail." The sokuon, that small tsu character, also extends into transliteration: Bach is written with it as Bahha, and Mach as Mahha, approximating sounds that have no native Japanese equivalent. Extended katakana digraphs were standardized in part by the Japanese Ministry of Education, Culture, Sports, Science and Technology, and some combinations were suggested separately by the American National Standards Institute and the British Standards Institution.

• 5. Katakana Beyond Japanese

  ›

  Japanese linguists use katakana to write the Ainu language, an indigenous language of northern Japan. Ainu has consonants at the ends of syllables, which standard katakana was not built to handle; the solution is a small version of a katakana character placed after the main syllable character. The Ainu word "up" is written ウㇷ゚, using a small pu to close the syllable. Ainu also requires three special modified characters using the handakuten diacritic. In Unicode, the Katakana Phonetic Extensions block, U+31F0 through U+31FF, exists specifically for Ainu. While Taiwan was under Japanese rule, a katakana-based writing system was developed to represent Taiwanese Hokkien, known as Taiwanese kana. It functioned as a phonetic guide alongside Chinese characters, similar in spirit to furigana in Japanese. Unlike the gojūon system, Taiwanese kana followed a methodology closer to bopomofo, with kana serving as initials, vowel medials, and consonant finals, and tonal marks added above. A dot below a kana marked an aspirated consonant. Katakana is also used as a phonetic guide for the Okinawan language; this system was devised by the Okinawa Center of Language Study of the University of the Ryukyus, and relies heavily on extensions and yōon digraphs to capture sounds that do not occur in standard Japanese.

• 6. Origins in Nara

  ›

  Buddhist monks in Nara created katakana in the 9th century, during the early Heian period. Their task was practical: they needed to transliterate texts and works arriving from India and China into something a Japanese reader could pronounce. Their solution was to lift fragments from existing man'yōgana characters, a system of Chinese characters borrowed for their phonetic value. Each katakana character descended from one portion of a Chinese character; the red markings often cited in historical tables show exactly which stroke or element survived into the modern symbol. Because the monks selected specific fragments consistently, the resulting script stabilized early. Katakana's choices of source segments were largely settled long before the Japanese government regularized scripts in 1900. Hiragana, by contrast, has more variant forms, because it evolved through a different process of cursive simplification. The linguistic scholar Eleanor Harz Jorden, described in the source as an influential American linguistics scholar, used katakana as the entry point in her textbook Japanese: The Written Language, paired with her spoken language course. Her approach introduced katakana before hiragana, on the grounds that loanwords give new learners meaningful material to practice reading immediately.

• 7. Katakana in Computing

  ›

  Katakana was added to the Unicode Standard in October 1991, with the release of version 1.0. The full-width katakana block occupies U+30A0 through U+30FF. Encoded alongside the characters themselves are the nakaguro word-separation dot, the chōonpu vowel extender, iteration marks, and a ligature of コト used in vertical writing. Half-width katakana, which fit into the same narrow pixel rectangle as Roman letters, originated with the JIS X 0201 encoding. In that scheme, diacritics were separate characters and each katakana was represented by a single byte, matching the capabilities of hardware at the time. In the late 1970s, two-byte character sets such as JIS X 0208 introduced full-width Japanese characters. For backward compatibility, both forms persist in Unicode; half-width katakana live in the Halfwidth and Fullwidth Forms block, starting at U+FF65 and ending at U+FF9F. Although sometimes described as obsolete, half-width katakana remain in active use: mini disc titles accept only ASCII or half-width katakana, and they appear on computerized cash register displays, shop receipts, and Japanese digital television and DVD subtitles. The Unicode standard added Ainu phonetic extensions in March 2002 with version 3.2, historic kana forms in October 2010 with version 6.0, and the Kana Extended-B block containing Taiwanese kana in a later release. The circled katakana characters, code points U+32D0 through U+32FE, appear in the Enclosed CJK Letters and Months block, with one conspicuous gap: a circled ン (n) is not included.