A rule for (some) Chinese tones

April 12, 2010

I have just come across an interesting rule for the tones of some words on a Popup Chinese podcast. The rule was only mentioned in passing, perhaps because it only applies to a small number of words. The rule is for when a single character can represent a noun or a related verb, which have the same pronunciation, but a different tone:

The verb is third tone, while the noun is forth tone

It's perhaps wrong to call this a rule, since it only applies to a few words, and for all I know there may be counter-examples. However, I was surprised that there was any such regularity in the Chinese tones; the example given was a word with which I have often confused the tones:

  • 处 chǔ - to reside, to live (also to get along with)
  • 处 chù - place, bureau (also respect)


  • 数 shǔ - to count, (also to criticize)
  • 数 shù - number, figure, several

Since hearing this 'rule', I have only been able to think of one more possible example and I'm not sure you can really separate the words into noun and verb:

  • 假 jiǎ - fake, to borrow, suppose
  • 假 jià - vacation

Discovering that such a rule exists makes me wonder whether there are similar rules for other words that I often confuse. I looked for more characters with multiple tones and it seems that most have first and forth tones):

  • 差 chā - difference, to differ, error, to err
  • 差 chà - differ from, short of, to lack

  • 发 fā - send out, develop
  • 发 fà - hair

  • 间 jiān - between
  • 间 jiàn - gap, to separate

  • 空 kōng - air, sky, empty, free time
  • 空 kòng - emptied, leisure

  • 要 yāo - demand, ask, coerce
  • 要 yào - important, to want, going to (as future auxiliary), may, must

I've also come across this pair, which has a different tone combination (although the verb does have a third tone):

  • 强 qiáng - strong, violent
  • 强 qiǎng - to strive

I guess I will just have to learn the differences.

Comments (3)

glossika on Sept. 18, 2010, 7:49 p.m.

I remember reading a more in-depth linguistics article about this over 10 years ago, I don't remember exactly but the author may have been Margaret Chan and written in English.

We may share an interest in finding such patterns across languages. I'm wondering with your programming background you could put together a function that describes tone changes in Wu (spoken in Zhejiang), although there are much more complex and challenging examples like Pingyao dialect of Shanxi.

Have you ever considered tracking large semantic groups based on rhyme? Of course, the rhymes would have to be historically accurate and not coincide in the modern language(s), but a good example would be the concept of cycle/completion/round/reciprocality in words ending in -an (zhuan, wan, yuan, ban, man, fan, huan, etc), or parallel/opposite in words ending in -iang (xiang, qiang, jiang), etc.

I have some videos discussing related topics on youtube under the same username.

Peter on Sept. 20, 2010, 1:48 p.m.

Thanks for your interest. Unfortunately, my knowledge of Chinese languages is limited: I know nothing about Wu or Pingyao. However, if you have a digital data set for these languages and a specific problem you'd like answering, I might be able to help.

I wasn't aware of the rhyming patterns - it sounds interesting. I'm not really sure how it would be possible to track changes. It is definitely possible to get a list of all words with a given ending, but then I think it would require a manual approach to see if they had similar meanings. Another possibility, if I could get hold of a digital Chinese thesaurus would be to see whether synonyms for certain words are disproportionately more likely to have a common ending. It might work.

Aurelio on Dec. 5, 2017, 5:50 a.m.

Hi Peter,

A few years late to the party, but google brought me here looking for examples of derivation by tone change. The phenomenon you describe is better known as qu-sheng derivation (from the 去-tone class that gave rise to the modern 4th tone in Mandarin). Generally speaking, you take a base word and change it to qusheng to turn a verb into a noun or a noun into a verb, or a general meaning into a more specific one (to give but some examples). There are many more than the ones you mentioned, so it is actually a good principle to know. Some are false positives, though, caused by the simplification of Chinese characters - "to send" and "hair" are two separate words in traditional Chinese, written very differently.



Leave a comment

cancel reply