I am pretty sure that some of the spoken varieties of modern Sinitic languages include words, morphemes, and/or particles which don't have a set written form.

Now I'm aware that some words or morphemes can be written with more than one character and that arbitrary words, including borrowings from other languages can be written with characters by way of a kind of transliteration.

What I'm asking about is either morphemes that exist only in spoken varieties of any language or dialect of Chinese and either have no written form, or can only be written in ad-hoc ways using transliterations or other kinds of character substitution.

I'm mostly interested in words/morphemes native to the variety in which they're used but borrowings may also be of some interest.

  • If a script offered no possibility to write a certain word it would not be a script (eg. Bliss). Every script can be used to write any word of any language! When you combine it with language, you get a writing system which acquires certain rulesets. These graphotactics afford single notations for many words, but several possible ones for others. Orthography arbitrarily limits these to 1 “set written form”. For most written languages, this happens in a never-ending invisible hand process, others have accepted authorities which are naturally limited. So the answer is yes!
    – Crissov
    Commented Oct 21, 2014 at 13:18
  • Related, on Quora: qr.ae/7ICWWc Commented Jun 10, 2015 at 6:34

3 Answers 3


Certainly. Here is one of many LanguageLog posts on the topic, showing two signs which use respectively the Japanese kana の and the bopomofo symbol ㄟ, in both cases to represent a Taiwanese possessive particle [e].

  • But that's incorrect. The correct Chinese character exists and it is 的. Writing that as の or ㄟ would have been deliberate choices to distinguish themselves from Mandarin. Neither represent a case of Taiwanese words lacking Chinese characters.
    – Semaphore
    Commented Jun 11, 2014 at 8:13
  • 3
    Semaphore: that is one possible interpretation of the facts. See the comments to the LanguageLog posting for discussion about whether it is the only possible interpretation.
    – Colin Fine
    Commented Jun 11, 2014 at 10:16
  • My speculation on the motives of using の or ㄟ is indeed no more than a possible interpretation, if one that's considerably more likely than LanguageLog's (show the right sign to a 100 Taiwanese, and more than 99 of them will say its intended to be in Hokkien). My factual statement on the existence of a Chinese character for ê however, is not. Frankly I'm not sure what could possibly have given Victor Mair the idea that ê had no written form.
    – Semaphore
    Commented Jun 11, 2014 at 10:49
  • Please try not to post made up statistics to a forum intended for experts. Did you read the entire on-topic part of the thread? Victor is only one of the several people with interesting things to say on the topic. Unfortunately there is also quite a bit of noise to skip over from some bigoted Han cultural warrior )-: Commented Jun 11, 2014 at 23:50
  • の can be written as 之, esp. in place names Commented Jun 12, 2014 at 3:03

I think it is safe to say that if the language remain alive long enough, it'd be a matter of time until a character is fitted to a spoken word. Even if it didn't originally start with one (i.e., European or certain Japanese loanwords in Taiwanese), or if the original had been lost. This seems to be typically achieved by:

  1. Choosing a character with a similar sound or meaning in Standard Mandarin.
  2. Choosing a character that has the same pronunciation in the dialect/language.
  3. Make up a new character.

Which, given the lack of a unified authority for most such languages, means there's probably a ton of dialectal words that have been written in different ways. And that's compounded by the enormity of the existing range of Chinese characters. On the other hand, that's less that a word couldn't be written properly, and much more the result of an intense marginalisation of these dialects and languages in the past century.

(The below is taken from my answer to the thread in the Chinese Stack Exchange)

There's a few Taiwanese words where the proper Chinese character was unknown or disputed or perhaps non-existent. This, combined with long term marginalisation of Taiwanese as a respectable medium of communication, resulted in a bit of chaos. On the other hand, in 2009 the government made an effort to standardise the writing system, so there's now a designated official characters for the more common words at least.

Anyway, here are some words that apparently had no universally agreed upon correct character (prior to 2009, at least):

  • bē/buē: officially 袂, was/is written as 𣍐, 未, indicating a negative
  • gâu: officially , was/is written as 賢, 爻, meaning "skilled at"
  • buaih: officially 勿愛 squished together, was/is written as 無愛, 覅, meaning "dislike"
  • suí: officially 媠, was/is written as 美, 水, meaning "beautiful"
  • tsia̍h: officially 食, was/is written as 噍, 吃, meaning "to eat"
  • phah: officially 拍, was/is written as 打, 扑, meaning "to pat, slap"
  • beh/bueh: officially 欲, was/is written as 要, 卜, meaning "to want"
  • kah: officially 佮, was/is written as 及, 甲, 合, meaning "together, with"
  • : officially 毋, was/is written as 不, 呣, 唔, indicating a negative
  • tshù: officially 厝, was/is written as 茨、戍, meaning "house".

I was browsing a language-learning book in the local public library today and came across this note on page ten of Get Started in Cantonese published by Teach Yourself:

Not all the written forms of Cantonese have been standardized. For example, aa could be written as 呀,吖,or丫;lāa could be written as 啦,罅, or 喇.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.