Japanese Romanization Quick Reference
Rules for writing Japanese words in romaji with correct macrons and formatting.
Long Vowels - When to Use Macrons
USE Macron (ō, ū, ā, ē)
Long o (おう/オウ, おお/オオ):
- Tōkyō (東京)
- Kyōto (京都)
- Ōsaka (大阪)
- rōmaji (ローマ字)
- shōgun (将軍)
- daimyō (大名)
- kendō (剣道)
- judō (柔道)
- shōyu (醤油)
- tōfu (豆腐)
Long u (うう/ウウ, う extending vowel):
- shūmatsu (週末)
- gakkō (学校)
- ryūgaku (留学)
- Ryūkyū (琉球)
- sūji (数字)
Long a (ああ/アア):
- okāsan (お母さん)
- obāsan (おばあさん)
Long e (ええ/エエ, えい sometimes):
- onēsan (お姉さん)
- sēfuku (制服) - but see exceptions below
DO NOT Use Macron
Long ii - always double i:
- Niigata (新潟) - not Nīgata
- oniisan (お兄さん) - not onīsan
- ojiisan (おじいさん) - not ojīsan
- chiisai (小さい) - not chīsai
Morpheme boundaries (meaningful units meeting):
- koushi (子牛) - ko + ushi, not kōshi
- otouto (弟) - can also be otōto depending on context
Dictionary form verb endings (-ou, -uu):
- omou (思う) - not omō
- tou (問う) - not tō
- nuu (縫う) - not nū
Hyphens - When to Use
Geographic Terms (Jurisdictions)
| Suffix | Example |
|---|---|
| -to (都) | Tōkyō-to |
| -fu (府) | Ōsaka-fu, Kyōto-fu |
| -ken (県) | Aichi-ken, Niigata-ken |
| -shi (市) | Yokohama-shi, Nagoya-shi |
| -ku (区) | Shibuya-ku, Chiyoda-ku |
| -chō/machi (町) | Ogawa-machi, Yūraku-chō |
| -mura/son (村) | Ogasawara-mura |
| -eki (駅) | Tōkyō-eki, Shibuya-eki |
Exception: Hokkaidō - always one word, no hyphen
Geographic features - NO hyphen:
- Fujisan (富士山)
- Sumidagawa (隅田川)
- Biwako (琵琶湖)
Titles and Honorifics After Names
| Suffix | Example |
|---|---|
| -san | Tanaka-san |
| -sama | Tanaka-sama |
| -kun | Nakamura-kun |
| -chan | Non-chan |
| -sensei | Yamamoto-sensei (or Yamamoto Sensei) |
| -shi (氏) | Haido-shi |
| -ke (家) | Tokugawa-ke |
Special Suffixes
| Suffix | Example |
|---|---|
| -teki (的) | Nihon-teki (日本的) |
| -gata (型) | Honkon-gata (香港型) |
| -shiki (式) | Tanaka-shiki |
| -ryū (流) | Urasenke-ryū |
| -sei (製) | Nihon-sei |
| -san (産) | Shizuoka-san |
Phonetic Changes in Compounds
When combining causes sound changes:
- kabushiki-gaisha (株式会社)
- fūfu-genka (夫婦喧嘩)
- Yamato-damashii (大和魂)
Word Division
Write as ONE Word
Common compounds:
- ukiyoe (浮世絵)
- origami (折り紙)
- sashimi (刺身)
- tempura (天ぷら)
- karate (空手)
- samurai (侍)
- tsunami (津波)
- sake (酒)
Prefixes that attach:
- daimyō (大名)
- shinkansen (新幹線)
- otanjōbi (お誕生日)
Suffixes that attach:
- keizaiteki (経済的)
- seibutsugaku (生物学)
- Bukkyō (仏教)
Write as SEPARATE Words
Modifiers with independent meaning:
- waga kuni (我が国)
- kono machi (この町)
Titles before names:
- Meiji Tennō (明治天皇)
- Sei Sabieru (聖サビエル)
Prefixes with proper names:
- Dai Nippon Teikoku (大日本帝国)
- ko Kurosawa Akira (故黒澤明)
Apostrophe
Use between n + vowel or n + y:
- shin’yō (信用)
- san’en (三円)
- hon’yaku (翻訳)
- Ken’ichi (健一)
- Shin’etsu (信越)
- Man’yōshū (万葉集)
Capitalization
Always Capitalize
Personal names - each word:
- Sugawara no Takasue no Musume
- Tanaka Tarō
- Kōbō Daishi
Geographic names - each word:
- Nihon Rettō (日本列島)
- Taiheiyō (太平洋)
- Bōsō Hantō (房総半島)
Corporate/organization names - each word except particles:
- Sensō o Kirokusuru Kai
- Kokusai Bukkyō Bunka Gakujutsu Kaigi
Religions and sects:
- Bukkyō, Shintō, Kirisutokyō
- Zenshū, Jōdo Shinshū
Historical/cultural events - each word:
- Dainiji Sekai Taisen (第二次世界大戦)
- Sekigahara no Tatakai
- Meiji Ishin
- Kantō Daishinsai
Structures, buildings, roads:
- Takamatsuzuka Kofun
- Narita Kūkō
- Kōshū Kaidō
Capitalize First Word Only
Titles of works (books, laws, artworks):
- Tsurezuregusa (徒然草)
- Asahi shinbun (朝日新聞)
- Rōdō kumiaihō (労働組合法)
- Fugaku sanjūrokkei (富嶽三十六景)
Historical periods:
- Jōmon jidai (縄文時代)
- Heianchō (平安朝)
- Shōwaki (昭和期)
Never Capitalize
Particles - no, wa, o, e, ni, de, to, etc.:
- Hotaru no hikari (蛍の光)
- Sekigahara no Tatakai
Derivatives with common noun meaning:
- gofuku (呉服) - not Gofuku
- setomono (瀬戸物)
- rōmaji (ローマ字)
- bosuton baggu (ボストンバッグ)
和/漢/洋 compounds (unless referring to nation/language):
- washitsu (和室), yōshu (洋酒), kanpō (漢方)
- But: Wa-Ei jiten (和英辞典) - refers to languages
- But: Wakō (和寇) - refers to national affiliation
Derivatives of Proper Names - Capitalize
When used as proper names:
- Nihonjin (日本人)
- Nihongo (日本語)
- Eigo (英語)
- Nihonshi (日本史)
- Bukkyōto (仏教徒)
Common Words Reference
| Japanese | Correct | Incorrect |
|---|---|---|
| 東京 | Tōkyō | Tokyo |
| 京都 | Kyōto | Kyoto |
| 大阪 | Ōsaka | Osaka |
| 柔道 | jūdō | judo |
| 剣道 | kendō | kendo |
| 相撲 | sumō | sumo |
| 豆腐 | tōfu | tofu |
| 醤油 | shōyu | shoyu |
| 抹茶 | matcha | - |
| 侍 | samurai | - |
| 将軍 | shōgun | shogun |
| 大名 | daimyō | daimyo |
| 芸者 | geisha | - |
| 神道 | Shintō | Shinto |
| 仏教 | Bukkyō | - |
| 茶道 | sadō/chadō | - |
| 華道 | kadō | - |
| 武士道 | bushidō | bushido |
| 空手 | karate | - |
| 合気道 | aikidō | aikido |
Country/Language Abbreviations
Hyphenate when combining:
- Nichi-Ro (日露) - Japan-Russia
- Wa-Ei (和英) - Japanese-English
- Nichi-Bei (日米) - Japan-US
But contractions are one word:
- Nitchū (日中) - Japan-China
- Nikka (日華) - Japan-China (historical)
Technical Identifiers in Code and Config Files
CRITICAL RULE: Proper romanization and capitalization ALWAYS take priority over programming naming conventions.
Apply Romanization to ALL Contexts
Use correct macrons and capitalization in:
- JSON keys and values
- Variable names
- Attribute identifiers
- Database field names
- API parameters
- File names
Examples
Correct approach:
attribute_name: "Shōgun-rank" (not "shogun-rank")
category: "Ryū" (not "Ryu")
location_id: "Tōkyō-station" (not "tokyo-station")
user_title: "Daimyō" (not "Daimyo")
Even in kebab-case or snake_case:
skill-type: "Kendō" ✓ (not "kendo")
power_source: "Shintō" ✓ (not "shinto")
npc-class: "Ōkami" ✓ (not "okami")
Capitalization Priority
Follow romanization capitalization rules, not code conventions:
- Proper nouns stay capitalized: Use
"Nihonjin"not"nihonjin" - Category names capitalized: Use
"Bukkyō"not"bukkyo"for religion types - Common nouns lowercase: Use
"sake"for the drink,"samurai"for warriors
Don’t Sacrifice Accuracy for Convention
Wrong thinking: “It’s a variable name, so make it lowercase” Right thinking: “It’s a Japanese proper noun, so capitalize it correctly”
Programming style guides bow to linguistic accuracy when dealing with non-English terms that have established romanization standards.