There seems to be a set of Chinese words that have alternatives (disregarding the traditional/simplified split). For example, and . Within this class of alternatives, however, seems to be a set of semantic puns. Consider, for example:

(1) (/su1/ 'revive') = 更生('live again') → (/su1/ 'revive')
(2) (/tian1/ 'sky') = 青色的空氣('blue air') → (/tian1/ 'heaven')

My question: is there a special term for these kinds of alternatives? Ultimately, I'd like to be able to find other such pairs.

NB1: I'm aware that the alternative in (2) is restricted to a Taoist context.
NB2: I'm also aware of alternatives like and , but from what I can tell, this specific pair seems to have gone the other way -- the semantic was later supplanted by the phono-semantic .

  • Great question! I am curious to know the answer too. As a native speaker I have never been able to internalize the pronunciations for those characters, especially 甦. Every time I encounter one of those characters, I have to think for a second before reading out. It's not a familiarity issue because I don't have problem with other rare non-radical-phonetic characters that have radical with a strong phonetic 'hint'. I think there's something fundamentally incoherent within the composition of those characters.
    – NS.X.
    Nov 23, 2014 at 20:26
  • would 甭 fall into the same category?
    – meireikei
    Nov 23, 2014 at 21:23
  • I guess 甭 fits the description of "semantic pun", but it lacks a "twin". I suppose I should add in my original question, that I've already consulted the resources provided from here.
    – Seralt
    Nov 23, 2014 at 21:37
  • 1
    Do you know if the pun-hanzis are from different origins - not only those hanzis created by 武则天 ?
    – meireikei
    Nov 23, 2014 at 21:54
  • I didn't even know about the Chinese characters of Empress Wu, so thanks for the lead! However, based on what I've been able to find on 蘇/甦 and 天/靝, these pairs have history independent of 武則天's influence. (Although, I'd definitely include "empress Wu pairs" like 照/曌.)
    – Seralt
    Nov 23, 2014 at 22:50

2 Answers 2


According to 六書 (http://en.wikipedia.org/wiki/Chinese_character_classification)

There are two kinds of this 1. 轉注: Basically this mean the multiple characters may map in the same mapping if 1) They are in same radical (category). 2) They sound similar Example: 老 (old) 考 (Originally meant old; deconstructed as 耂 derivative (老字头) of 老 lǎo "old" + 丂 qiǎo (phonetic). Later evolved to mean "deceased father". Modern meanings of to exam. (From http://en.wiktionary.org/wiki/%E8%80%83)

  1. 假借: Basically this mean the a character is assigned to more than one senses. Example: 蘇 (/su1/ 'revive') = 更生('live again') → 甦 云 (Originally Cloud, then become "say") → 雲 (Cloud)

Or you can search "通同字" or "通假字"


What about 台 and 臺?

台 is both "simplified" and "alternative", 臺 is "traditional".

臺 seems to be the "correct" traditional form for Taiwan, but I've never seen it in reference to Taishan (台山) in Guangdong, even in all references pre-1949, etc.

How is it determined in "traditional" usage when to use the one or the other?

(In simplified usage 台 is just used all the time.)

  • Yeah, I would consider stuff like 台 and 臺 as a "fringe" case. Same for 才 and 纔. In both these cases, the history is clearly that of a traditional character being replaced by a simplied one in traditional script.
    – Seralt
    Dec 4, 2014 at 17:38

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.