Romaji: Meaning (information, definition, explanation, facts)

In Japanese, rōmaji (ローマ字 "Roman characters") broadly refers to the Roman alphabet.

In English, the word refers specifically to the writing of the Japanese language in Roman characters as opposed to the usual mix of kanji and kana.

Rōmaji is sometimes mispelled as Rōmanji, and although seen with some frequency, this spelling does not represent any Japanese spelling or pronunciation of the word.

Japanese may be written in rōmaji for many reasons: street signs for visiting foreigners; transcription of personal, company, or place names to be used in another language context; dictionaries and textbooks for learners of the language; or even simply for typographic emphasis.

There are a number of different romanization systems in use: the three main ones are Hepburn, Kunrei-shiki (Cabinet order or ISO 3602), and Nihon-shiki (ISO 3602 Strict). Hepburn (long-vowel omitted) is the most widely used. Modified Hepburn, which uses a macron to indicate some long vowels and an apostrophe to note the separation of easily confused syllables (for example, the name じゅんいちろう is written with the syllables jun-ichi-ro and u, and is romanized as Jun'ichirō in Modified Hepburn) is widely used in Eastern Japan and among foreign students and academics.

Differences between the systems

• Hepburn follows the phonology of Romanic languages and so gives the best indication to people familiar with one of those languages of how a word is pronounced in Japanese. It was standardized as American National Standard System for the Romanization of Japanese (Modified Hepburn), but this status was abolished on October 6, 1994. Hepburn is used by Wikipedia and is the most common romanization system in use today, especially in the English-speaking world. Japanese school children now learn Hepburn when they first begin to learn the English alphabet in junior high school.

• Nihon-shiki The sole difference between Nihon-shiki and Kunrei-shiki is a relic of the differences between the kana syllabary and pronunciation. For example, if the words kana かな and tsukai つかい are combined, in kana the result is written かなづかい with a dakuten (voicing sign) ゛ atop to indicate that the tsu つ is now voiced. Hepburn and Kunrei-shiki ignore the underlying kana and represent the sounds as they are pronounced ("kanazukai"), but Nihon-shiki retains the difference and romanizes the word as "kanadukai". This makes Nihon-shiki one of the few systems of romanization that allows lossless mapping to and from kana. It has also been known as ISO 3602 Strict form.

• Kunrei-shiki is a modified Nihon-shiki with a touch of Hepburn; it has been standardized by the Japanese Government and ISO (ISO 3602). Kunrei-shiki used to be taught to all Japanese elementary school students.

In addition to the standardized systems above, one can see many other romanizations. These are used by many people, either because they do not fully understand the particular system they are attempting to use, or for deliberate stylistic reasons. Macrons and other diacritical symbols are often omitted or substituted for, because of carelessness, difficulty in remembering or inputting them, or simply unavailability in one's character set (although this last reason is becoming less frequent with the widespread introduction of Unicode).

Also commonly seen are wāpuro rōmaji, referring to the various methods that IMEs use for converting keystrokes on a roman keyboard to kana. (Wāpuro derives from wādo purosessā [word processor].) Unlike the standard systems, wāpuro rōmaji requires no characters from outside the ASCII character set.

Romanizations that one is likely to come across "in the wild" include:

• oh for おお or おう (Hepburn ō).
• ou for おう (also Hepburn ō). This is an example of wāpuro rōmaji. This romanization style is popular with fans of anime (otaku).
• ô for おお or おう (Hepburn ō). This is valid Kunrei-shiki, but occasionally occurs in otherwise Hepburn-romanized words due to confusion or substitution (since ô exists in ISO-8859-1 but ō does not).
• jya for じゃ (Hepburn ja) and so on. This seems to be the result of confusion between the Hepburn and the other romanization systems.
• dzu for づ (Hepburn zu). Another combination between multiple systems, in this case Hepburn and Nihon-shiki.
• cchi for っち (Hepburn tchi) and so on. This is often used for stylistic reasons when rendering nicknames (e.g., あきこ Akiko becoming あっちゃん Acchan rather than Atchan).
• la for ら (Hepburn ra) and so on. Since the Japanese consonant /r/ has a sound (IPA ɽ) that is somewhat between an English "r" and an "l" (and to some listeners sounds somewhat like a English "d"), this is unsurprising.
• a for ああ (Hepburn ā) and so on -- in other words, merely failing to mark long vowels at all.
• na for んあ (Hepburn n'a) and so on.
• nn for ん (Hepburn n). This is also an example of wāpuro rōmaji, although many IMEs also accept the Hepburn n'. Since this leads to ambiguity with the more widespread Hepburn system (in, e.g., the cluster nna, which unambiguously represents んな in Hepburn but would be んあ in this system), this form occurs only rarely.

While there may be arguments in favour of these romanizations in specific contexts, their use (especially if mixed) generally leads to even greater confusion, especially when Japanese words are romanized for indexing in a database.

Personal names can be subject to even more variation, with spellings depending on the individual's preference. For example, the manga artist Yasuhiro Nightow's family name would be more conventionally written in Hepburn as Naitō.

In addition, words and names that have established English spellings, such as Tokyo, Kyoto, and Kanto, are sometimes written as they are in English, without regard for the rules of romanization.