Hepburn vs Kunrei-shiki Romanization: Systems, Rules and Examples

Native Writing System

Japanese writing combines three scripts: kanji (logographic characters borrowed from Chinese), hiragana and katakana (two syllabaries of 46 basic symbols each). Hiragana represents native grammatical elements, while katakana is used for foreign loanwords, onomatopoeia, and emphasis. The phonemic inventory of Japanese is relatively small, consisting of five vowels and a set of consonant‑vowel (CV) syllables.

Major Romanization Systems

Yoshio Mikami (University of Tokyo) notes, “Consistent romanization is essential for reliable data exchange, especially when Japanese names appear in international databases.”

Recommended System by Use Case

For an English‑speaking audience, Hepburn is the most widely recognized system. It balances phonetic readability with scholarly precision, making it suitable for:

• Academic papers: journals in Japanese studies and comparative linguistics prefer Hepburn because it conveys pronunciation to non‑Japanese readers.
• Travel guides and signage: tourists benefit from spellings that approximate English sounds (e.g., Kyōto rather than Kyoudo).
• Digital platforms: search engines and social media recognize Hepburn forms as the standard index terms.

Kunrei‑shiki remains the official system for Japanese government publications and elementary education, but its spellings can be unintuitive for English speakers (e.g., si for shi).

Alphabet or Character Chart

Vowels

Japanese has five pure vowels. Long vowels are indicated in Hepburn with a macron (ā, ī, ū, ē, ō) or by doubling the vowel in Kunrei‑shiki.

• a – あ, ああ, あー → a / aː
• i – い, いい, いー → i / iː
• u – う, うう, うー → u / uː
• e – え, ええ, えー → e / eː
• o – お, おお, おー → o / oː

In rapid speech, /u/ often becomes devoiced, a point that romanization cannot capture.

Consonants

Consonants appear only before a vowel (or the moraic n). Key points:

• k, g, s, z, t, d, n, h, b, p, m, y, r, w map directly.
• Palatalized sounds: shi (し) and chi (ち) are rendered as “shi” and “chi” in Hepburn, but as “si” and “ti” in Kunrei‑shiki.
• The nasal n (ん) is written “n” before consonants, but becomes “m” before bilabial sounds (e.g., senpai → “senpai”, senmitsu → “senmitsu”).

Diacritics and Tone Marks

Japanese does not use lexical tone; however, pitch accent is phonemic and cannot be expressed with standard romanization.

Word Separation

Native Japanese is written without spaces. In romanized text, spaces are inserted between lexical items to aid readability, especially in academic and travel contexts. Example: 「日本語を勉強します」 becomes nihongo o benkyō shimasu.

Capitalization

Capital letters follow English conventions:

• Sentence‑initial words are capitalized.
• Proper nouns (people, places, organizations) retain initial capitals (e.g., Tokyo, Yamamoto).
• All‑caps are avoided except for abbreviations.

Names

Japanese personal names are traditionally family‑given order. International publications often invert the order to given‑family for consistency with Western norms. Hepburn romanization renders long vowels with macrons unless the publication prefers the double‑vowel style.

Example:

• 山田 太郎 → Yamada Tarō (family‑given)
• Tarō Yamada (given‑family, used in passports)

Place Names

Geographic names follow the same Hepburn conventions used in travel guides. Historical spellings (e.g., “Edo” for 東京) are retained only in specialized contexts.

Examples:

• 京都 → Kyōto
• 大阪 → Ōsaka
• 広島 → Hiroshima

Examples

Three real‑world sentences illustrate the recommended Hepburn output and a brief pronunciation guide (IPA in brackets).

1. Japanese: 今日は東京へ行きます。
   Hepburn: Konnichiwa Tōkyō e ikimasu.
   Pronunciation: [konnit͡ɕiɰa toːkʲoː e i̥kʲimasɯ]
2. Japanese: この本は面白いです。
   Hepburn: Kono hon wa omoshiroi desu.
   Pronunciation: [kono hoɴ wa omoɕiɾoj̥e desɯ]
3. Japanese: 明日は雨が降ります。
   Hepburn: Ashita wa ame ga furi masu.
   Pronunciation: [aɕita wa a̠me ŋa ɸɯɾi masɯ]

Alternative Spellings

Because Kunrei‑shiki preserves kana morphology, many words have two accepted spellings:

• shi (Hepburn) ↔ si (Kunrei‑shiki)
• chi ↔ ti
• tsu ↔ tu
• Long ō ↔ ou (e.g., Tōkyō ↔ Toukyou)

When consistency is required, choose the system that matches your audience’s expectations.

Pronunciation Limitations

Romanization cannot convey:

• Pitch‑accent patterns, which differentiate words such as hashi (bridge) vs. hashi (chopsticks).
• Allophonic devoicing of /u/ and /i/ in unaccented syllables.
• Geminate consonants indicated only by the small tsu (っ); Hepburn marks them with a doubled consonant (e.g., kippu for きっぷ).

Work‑arounds include adding IPA transcriptions in parentheses or using diacritics to flag pitch when needed for linguistic work.

Converter

Several reliable online tools automate Hepburn and Kunrei‑shiki conversion:

• Hepburn Converter – simple paste‑and‑convert, supports macrons.
• Kunrei‑shiki Converter – follows ISO 3602‑2 standards, useful for government forms.
• Nihon‑shiki Converter – one‑to‑one kana mapping for reversible transliteration.
• Multi‑System Converter – lets you switch between all three systems in one interface.

For bulk processing, the Japanese Romanization API (available via the same domain) accepts UTF‑8 input and returns JSON with selected system output.