Man'yōgana: the early phonetic use of Chinese characters in Japanese

Characteristics and structure

Man'yōgana is fundamentally phonetic: each character stood in for a mora or syllable of Japanese speech. Because kanji carry multiple historical readings, and because Chinese and Japanese phonologies differed, many different characters could be used to write the same Japanese sound. At its height the repertoire included several hundred characters used in this manner to represent a far smaller set of syllables or morae. Early lists and analyses estimate that roughly nine dozen phonetic units of Old Japanese could be expressed in man'yōgana, and the number of distinct characters used to represent them ran into the high hundreds. The practical result was a non‑standardized but flexible orthography in which the same word might appear written with different characters in different texts, or even within a single composition. reference 2

The system made a basic distinction that later became central to Japanese orthography: a character could be used either for its meaning (semantic use, what Japanese later called kanji in everyday sense) or for its sound (phonetic use, man'yōgana). Semantic use typically preserved the Chinese concept attached to the character; phonetic use ignored semantic content and treated the character as a label for a sound. Writers often mixed both uses in the same sentence: for example, a verb stem might be written with a semantic character while its inflectional ending was expressed by man'yōgana. The mixed pattern could make texts compact but also hard to read for later generations unaccustomed to the conventions. reference 3

Historical background and development

The arrival of Chinese writing in Japan, via the Korean peninsula, furnished the raw material for literate culture in Japan from roughly the fourth to sixth centuries CE onward. Initially, Chinese characters were learned and used to write classical Chinese, a prestige written language that served government, ritual and scholarship. Over time Japanese scholars and clerks adapted those characters to the local language. One early and practical response — man'yōgana — exploited the phonetic possibilities already present in the Chinese script to capture Japanese sounds that did not correspond neatly to Chinese morphemes or sentence structure. reference 4

By the end of the eighth and into the ninth centuries new shorthand solutions began to supplant man'yōgana in everyday use. Two cursive traditions emerged from the practice of writing characters in abbreviated or fragmentary ways: a flowing, cursive script favored by court women evolved into hiragana; and clipped fragments or linear elements extracted from characters by Buddhist monks gave rise to katakana. These kana scripts encoded individual morae in standardized single signs rather than whole characters, making writing faster and reading clearer. The transition was gradual: manuscripts of the ninth and later centuries show mixed use of kanji, man'yōgana and early kana, and the full establishment of kana as the standard phonetic complement to kanji took several generations. reference 5

Examples, uses and legacy

Man'yōgana is most famously associated with the Man'yōshū, the earliest large collection of Japanese poetry compiled in the Nara period. Poets used man'yōgana to record vernacular forms, inflections and native lexemes that could not be handled by classical Chinese alone. A simple modern sentence such as 私はすしを食べます (Watashi wa sushi o tabemasu, "I eat sushi") can be written in mixed kanji and kana, but when authors attempted to render every syllable with Chinese characters the result took on a markedly different appearance: known examples from scholarly reconstruction show how familiar morphemes might be represented by characters chosen for pronunciation rather than meaning. The visual consequence was a text that looks superficially like Classical Chinese to a modern reader but that must be read with Japanese phonology and grammar in mind. reference 6

• Phonetic examples: a native word such as yama ("mountain") could be written by characters chosen for the syllables ya and ma; the semantic character 山 continued to be used for the meaning ‘mountain’ especially in Sino-Japanese compounds.
• Ateji: some loanwords and proper names continued to be written with characters selected for sound regardless of meaning; later corpora show conventional ateji spellings like 寿司 for sushi. Such spellings survive as archaic or stylistic forms in modern usage. reference 7
• Orthographic diversity: multiple characters could represent the same mora, reflecting regional, chronological or authorial preferences.

Because man'yōgana used whole characters, texts written mainly in this system tended to be dense and time-consuming to write by hand; a single syllable could take many strokes. The shift to kana significantly reduced labor and increased literacy among non‑elite readers and writers. Nonetheless man'yōgana left durable traces: the shapes of hiragana and katakana signs are directly derivable from specific man'yōgana spellings, and early glosses and annotations in Buddhist and court texts preserve man'yōgana spellings that are valuable to modern historical linguists reconstructing Old Japanese phonology. Readers interested in paleography and historical phonology can compare surviving man'yōgana examples to later kana forms to trace those developments. reference 8

Distinctive features, scholarly significance and notable facts

Several features make man'yōgana important to students of Japanese language history. First, because writers often employed different Chinese readings to represent native sounds, man'yōgana supplies evidence for distinctions in Old Japanese vowel qualities and consonant clusters that later merged. Scholars have used man'yōgana spellings to infer distinctions sometimes called the ancient kana usage (e.g. Jōdai Tokushu Kanazukai), which illuminate a stage of Japanese sound structure lost in modern standard pronunciation. reference 9

Second, the practice demonstrates early attitudes to literacy and language: classical Chinese was a learned, elite medium, while man'yōgana allowed vernacular speech and poetic expression to be recorded. The split anticipates later distinctions in written Japanese between kanji (for lexical roots and Sino-Japanese vocabulary) and kana (for inflectional endings, particles and native words). Third, man'yōgana explains why many modern kana signs have multiple historical origins: a single kana may consolidate the use of several different characters that earlier writers had used for the same syllable. reference 10

Finally, traces of man'yōgana linger in modern practices. Words sometimes appear in present‑day Japanese with kanji that function mainly as phonetic or associative signs rather than precise semantic descriptors; such usage — often labelled ateji — is visible in historical spellings of foreign names and brand or place names. Examples often cited in textbooks and popular discussions include conventional character spellings for foreign place names and loanwords preserved from earlier orthographic habits. Where contemporary writing uses katakana for foreign terms, older ateji forms reflect the intermediate steps between man'yōgana and modern loanword conventions. reference 11

Practical distinctions and reading challenges

For modern readers, man'yōgana presents two main challenges. One is ambiguity: because a character used phonetically contributes no semantic cue, passages in which many syllables are written phonetically can be hard to parse. The other is orthographic variation: the absence of a single standardized inventory means that a reader must often supply context or specialized knowledge to identify which syllable a given character represents. Japan's later development of standardized kana scripts deliberately addressed both problems by creating a one‑to‑one mapping from sound to sign, and by reserving kanji primarily for lexical meaning. Yet the older habit of mixing phonetic and semantic uses survives in certain literate practices and in scholarly editions of early texts, and scholars continue to study man'yōgana to learn about phonology, orthography and the social contexts of writing in early Japan. reference 12

Resources for further study

Readers seeking primary sources will find man'yōgana examples in edited collections of the Man'yōshū and in early inscriptions and administrative documents; specialized studies in historical linguistics discuss how man'yōgana informs reconstructions of Old Japanese vowels and consonant distinctions. Introductory treatments of Japanese writing systems typically situate man'yōgana as the bridge between classical Chinese literacy and the kana systems. For concise historical overviews, general works on East Asian scripts and histories of Japanese literature are useful. reference 13 reference 14

• Primary literary witness: Man'yōshū (examples of man'yōgana spellings and poetic usage). reference 15
• Script evolution: studies on the origin of hiragana and katakana from cursive and fragmentary character forms. reference 16
• Phonology: analyses that use man'yōgana to infer Old Japanese sound contrasts (e.g., distinctions recorded in ancient kana usage). reference 17

For readers curious about visual examples: facsimiles and diplomatic transcriptions of early manuscripts show how scribes mixed character choices and how the forms of particular kanji were altered or simplified in phonetic use. These representations are commonly reproduced in academic introductions and in museum catalogues, where high-resolution images and careful transcription conventions help non‑specialists compare man'yōgana spellings with later kana forms. reference 18

Man'yōgana is therefore both a historical orthographic technique and a key to understanding the emergence of Japan's mixed writing system. It demonstrates an early and creative adaptation of imported script to native linguistic needs, and it remains an important source for scholars reconstructing the sound system, literacy practices and literary expression of early Japan. Interested readers can consult general surveys of Japanese writing and more specialized linguistic treatments for detailed charts, examples and photographic reproductions of surviving man'yōgana texts. reference 19 reference 20 reference 21