The iteration marks 々, ヽ, ヾ, ゝ, ゞ are slightly smaller than standard kana, and are often seen in print when a contemporary scholar is transcribing text, a printmaker's seal, etc. My question relates to the kunojiten (くの字点, くのじてん), either the 〱 (unvoiced) or 〲 (voiced) repeat marks that resemble the hiragana characters ku (く) and gu (ぐ). Given they are only used in vertical text, stretching to fill the space usually occupied by two kanji characters, what is the best solution if you are TRANSCRIBING them from pre-20th century documents and cartouches, typing them up?

Perhaps this is a little bit of a style question... might one write:

摂州大物の浦よしつねしう〲難風... and risk a reader thinking 〲 is gu

摂州大物の浦よしつねしう〲[じう]難風... i.e., interpolate in square brackets

摂州大物の浦よしつねしう[じう]難風... interpolate in square brackets with no reference to 〲

or simply transcribe as

摂州大物の浦よしつねしうじう難風... with no reference to there having been a kunojiten present?

My reference from a cartouche: 摂州大物の浦よしつねしう[kunojiten]難風に出合て平家のぼうれい御船を覆さんとする圖

or with accompanying furigana in the cartouche: 摂州[せっしゅう]大物の浦よしつねしう[kunojiten]難[なん]風に出合て平家のぼうれい御船を覆[くつがえ]さんとする圖, Sesshū Daimotsu-no-ura Yoshitsune shiujiu [shūjū] nanpū ni deaite Heike no bōrei mifune wo kutsugae san to suru zu.

Essentially, is there a consistent and/or best practice for transcribing or transliterating kunojiten in printed text?

Shiujiu (しう〲, thus しうじう), appears to be a more archaic way to write shūjū (しゅうじゅう) or shujū (しゅじゅう), a "master and servant" or "lord and retainer." Shūjū/shujū is sometimes seen in kanji form (主従) but it's usually seen Shiujiu (しう followed by the kunojiten) in 19th century prints. The dictionaries I have access to (at present) don't bring up the spelling Shiujiu... not quite sure when it changed, and the historical relationship/shift of 'iu' to 'ū'.

  • 1
    As far as two vowels being together, that is not the case in your example. 主従 is しゅ(not a vowel) + じゅ(also not a vowel) + う (vowel). They are only vowels if they stand alone and are not a combination syllable (consonant + vowel).
    – BJCUAI
    Commented Mar 23, 2018 at 21:57
  • @user27280, thanks for your succinct and perfect comment, which reminds me that (i) I should not be diluting my question with too many notes, and (ii) shouldn't be 'googling' at 4 in the morning and cursorily trying to frame and hypothsise. Your notes on the vowels exposes my folly... and I have edited out the sentences in my question as a result (couldn't work out how to do a 'strikethrough')! Thank you... your comment stands with all its weight, and should not be too 'orphaned' even with this deletion.
    – musha
    Commented Mar 24, 2018 at 2:00
  • Glad that it helped. I'll keep this short, as it's 4:00 am.
    – BJCUAI
    Commented Mar 26, 2018 at 10:58

2 Answers 2


Aozora Bunko renders the くの字点 as a forward slash plus backward slash (/\ or /″\).

One option would be to use multiple single iteration marks (Ex. ところゞゝゝ, 馬鹿々々しい). In your case, it could be しうゞゝ. I think this would cause the least confusion.

Edit: Also, 主従{しゅうじゅう} being spelled しうじう reflects kana orthography that predates World War II. See the Wikipedia article Historical kana orthography for more details.

  • Hi @JAM, I would prefer to avoid single iteration marks that are not there in the original text... essentially, I am interested in transcribing the variant kanji/kana forms and iteration marks that are used in the original. In this regard, if the kunojiten 〲 runs the risk of being confused with gu ぐ, and other options are messy/deceptive, I would definitely use /″\ if that is considered current best practice. I notice that on web links looking at typographic symbols and unicode, a 〴 is placed directly above 〵 to mimic the kunojiten, but for that I need to type vertically, not horizontally!
    – musha
    Commented Mar 25, 2018 at 3:05
  • Further research: In a small preface, Herbert Jonnson ("Reading Japanese Haikai Poetry: A Study in the Polyphony of Rosa Buson's Linked Poems", 2016) says: "The extended repetition mark (kunojiten) used in the original, which can only be used in vertical script, has been replaced by the ordinary modern repetition mark (々) for the kanji and words in kana are written out in full without repetition marks." So that was his approach/solution! On a language forum an OP asks about Yumeno Kyuusaku poems and the constant use of /\, e.g. 可愛き小鳥やは/\と, so that was acceptable practice in that text.
    – musha
    Commented Mar 25, 2018 at 3:32

Two options for transcribing a handwritten source text with kunojiten into a typescript are:

  1. as JAM's answer suggests, use /\ and /″\ when typing horizontally 摂州大物の浦よしつねしう/″\難風に出合て平家のぼうれい御船を覆さんとする圖

  2. use 〳 and 〵, or 〴 and 〵 placed directly above each other when typing vertically. This works well in a Microsoft Word doc (create a one cell table, and change text direction). As soon as 〴 and 〵 are typed in they form a double-character length 〲.

vertical script with a kunojiten

This example (above) doesn't account for the furigana, but it's indicative. Original text below:


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .