Skip to the text
Parseh
پParseh guide The Markdown dialect › Documents in each language›Writing Japanese
On this page

Writing Japanese

target: ja — Japanese, 日本語: kanji, hiragana and katakana, written left to right, or in vertical columns read from right to left. It is the one target language with a reading, the kana over the kanji.

1. How its text is found#

Japanese needs no mark. Kanji, hiragana, katakana, the Japanese punctuation 。、「」 and the full-width signs are found in the prose on their own. Japanese puts no spaces between its words, so a run is the characters that touch: a space, a Latin letter or a western digit ends it.

Parseh Markdown
---target: ja---Japanese needs no mark: 日本語 *nihongo*, and 水 = *water*.猿も木から落ちる。毎朝7時に起きて、8時に家を出ます。[週末はDVDで映画を見ます。]{tl}
Result

Japanese needs no mark: 日本語 nihongo, and 水 = water.

猿も木から落ちる。

毎朝7時に起きて、8時に家を出ます。

週末はDVDで映画を見ます。

  • A paragraph of nothing but Japanese characters and Japanese punctuation is a display line, set larger.
  • A paragraph with no Latin letter — western digits allowed — is a Japanese paragraph.
  • A word or a sentence with Latin letters or digits inside it is marked […]{tl} or […]{ja} to stay one: [Tシャツ]{tl}, [3時]{tl}.

Keep each sentence on one line of the source. The studio joins the lines of a paragraph with a space, which Japanese does not want.

2. Faces#

On the screen Japanese is set in the Noto Serif CJK JP of your device, or its mincho face (Hiragino Mincho, Yu Mincho, IPAex Mincho, MS Mincho) — no Japanese face travels with the toolbox; on paper in Noto Serif CJK JP. The second face, font=gothic, is Noto Sans CJK JP, the plain sans-serif of signs and screens. Japanese text is set 1.20 times the size of the prose.

Parseh Markdown
---target: ja---[本日休業]{tl font=gothic bg=rose}
Result

本日休業

3. Vertical columns#

vertical — or tategaki, or mode=vertical — sets a block in columns, top to bottom, the columns following each other from right to left, as books and poems are set. height= is a column’s height in em, from 8 to 60, 22 when left out; on the screen a block wider than the page scrolls sideways. A vertical block is always a whole paragraph.

Parseh Markdown
---target: ja---[古池や⏎蛙飛び込む⏎水の音]{tl vertical height=8}[The old pond / a frog jumps in / the sound of the water.]{la align=center bg=sand}
Result

古池や
蛙飛び込む
水の音

The old pond / a frog jumps in / the sound of the water.
Parseh Markdown
---target: ja---[菜の花や⏎月は東に⏎日は西に]{tl tategaki height=10}
Result

菜の花や
月は東に
日は西に

On paper the columns are set rotated — true vertical setting — and the block’s tint is left out; in large print a column is never longer than the page.

4. The reading#

A word may carry its reading in kana beside its rōmaji: kana: (or reading:) and translit:, in either order, a colour before them.

Parseh Markdown
---target: ja---[漢字]{kana:かんじ} has its reading, [東京]{translit:Tōkyō} its rōmaji,[日本語]{kana:にほんご translit:nihongo} both, and[学校]{indigo kana:がっこう translit:gakkō} a colour as well, and[先生]{reading:せんせい translit:sensei} uses the other name of `kana:`.
Result

漢字 has its reading, 東京 its rōmaji, 日本語 both, and 学校 a colour as well, and 先生 uses the other name of kana:.

Pointing at the word shows the kana above the rōmaji; the glossary has a reading column; on paper the word is printed alone. In a Japanese document the editor shows a kana button, which wraps the selection in […]{kana:}, and the reading view’s cloud has a field for the kana beside the one for the rōmaji.

5. Vocabulary entries#

Four parts after the headword, the reading first: kana, rōmaji, origin, meaning. The second part is always the reading; leave it empty (## 猫 | | neko) to give only the rōmaji.

Parseh Markdown
---target: ja---## 食べる | たべる | taberu | = *to eat*## [猫]{teal translit:neko} | ねこ | = *cat*
Result
食べる
たべる
taberu
猫
ねこ
neko

6. Rōmaji#

Hepburn, written from the reading, the same in every document whatever the prose is written in:

  • long vowels with macrons, ō ū (Tōkyō, kūki); ei and ii as written (sensei);
  • the particles は, へ, を as they sound: wa, e, o;
  • っ doubles the next consonant (gakkō, kitte; matcha);
  • ん is n, and n’ before a vowel or y (kin’en); never m (shinbun);
  • shi chi tsu fu ji, sha shu sho, cha chu cho, ja ju jo;
  • words parted where a dictionary parts them, a particle as a word of its own, a verb with its auxiliaries (tabeteimasu); capitals only for names.

In the kana of a reading, particles are written as written (は, not わ), long vowels as spelled (とうきょう), and katakana stays katakana.