When I started learning Chinese I would often hear (mostly northers) 连读 a bunch of different words:

意思 yi si - became - yis

认识 ren shi - became - rensh

Perhaps this doesn't count as 连读 - but the separate syllables where definitely combined into one distinctive syllable.

Any ideas?

  • I'm not sure I get your question. You believe that there's some sort of elision causing those words to be pronounced as one whole syllable?
    – blackgreen
    Commented Jan 4, 2017 at 8:02
  • @TXV Yes, exactly. In certain accents - from certain areas. It's not everyone who pronounces things this way - but this phenomenon certainly does exist.
    – Mou某
    Commented Jan 4, 2017 at 9:02
  • The only pattern I'm aware of that can relate to your question is that what you call 连读 word is comprised of one 4th tone syllable + one 5th (neutral) tone syllable. Neutral tone following a descending tone results in a 411 (or 511) pronunciation. Which makes the second character sound like a void syllable. The fact that you perceive an elision is due to the fact that the final "i" is by itself pronounced as a very closed vowel. This plus its low neutral tone makes it almost "disappear". Is 连读 the actual term for describing all this?
    – blackgreen
    Commented Jan 4, 2017 at 10:24
  • If I'm not mistaken, that final "i" should be represented with this IPA symbol /ɨ/
    – blackgreen
    Commented Jan 4, 2017 at 10:32
  • 連讀 in a phonetic context is equivalent to English liaison and in the context of Chinese linguistics is usually discussed with reference to tone sandhi. Changes involving consonants and vowels do exist in some topolects; all three are found in the Fuzhou dialect
    – Michaelyus
    Commented Feb 3, 2017 at 16:59

5 Answers 5


Some have been crystallized in the spelling: - 别 < 不要 (when used as a negative imperative) - 甭 < 不用 (it's even explicit in the character!) - 啦 < 了啊 - 这/那 as zhei/nei < 这一/那一 (zhe yi / na yi) There are probably others, plus all the Classical Chinese ones mentioned by Tang Ho.

When it comes to ones that have not been represented in writing: - A very common one I can think of is 多少钱, in which 多少 easily becomes "duo'ao" (or more likely, "dwao"). - 这样子 is also commonly pronounced as "zhiangzi". - 不要 is often "b'yao", which has a character 嫑 biao2 which I've absolutely never seen used

Some characters also suggest that the following contractions exist, although they seem to be dialectal and I've never seen/heard them used: - 嘦 < 只要 jiao4 - 覅 < 勿要 fiao4 - 孬 < 不要 nao1


In Cantonese, we have something called 'contraction' It is a rather common practice, if the second syllable involved has a coda.

For example:

C.Chiu wrote

一陣間 jat zan gaanja(t)(za)n gaanjan gaan 因間

[loss of coda /t/ and the following onset /z/ + vowel/a/]

地方 dei fongde(i) (fo)ngdeng

[loss of coda /i/ and the following onset /f/ + vowel /o/]

夠膽死 gau daam seiga(u) (da)am seigaam sei 減死(敢死-gam sei)

[loss of coda /u/ and the following onset /d/ + vowel /a/]

As for Mandarin, there's some noticeable contractions of classical Chinese

what are common contractions in classical chinese?

諸 = 之乎 (In some rarer cases 諸 can also be contraction for 有之乎. 諸 can be used on its own with the meaning of "all, the class of", as in 諸侯 "the feudal lords.") 焉 = 於之 (於之 is never used; only 焉.) 旃 = 之焉 (Rare.) 爰 = 于之 (Rare. The prepositions 於, 于, and 乎 are of different origin, but used interchangeably (except that 乎 can also be used as a final question particle).)

  • 然 = 如之
  • 云 = 曰之
  • 弗 = 不之
  • 勿 = 毋之 (弗 and 勿 were originally not contractions, but were reanalyzed as contractions in the Warring States period.)
  • 耳 = 而已
  • 盍 = 胡不 (胡 is a variant of 何.)
  • 與 = 也乎 (Also written 歟.)
  • 邪 = 也乎 (Also written 耶. Probably a dialectal variant of 與.)
  • 夫 = 不乎 (夫 has many other meanings.)
  • 奈何 = 若之何

I do not speak Mandarin, so I can't think of any modern Mandarin phrase contractions. But the examples from the post above proved contractions exist in Mandarin too.


One example where 连读 occurs is when a final -ng comes before a vowel, w, or y. For example: 中央 = zhong (zhung) yang = zhuang, 公安 gong (gung) an = guan. At least around Beijing

Some patterns also start to make sense when you see what vowels and dipthongs can disappear or reduce. e.g. ei > i, a, an, ang > e, en, eng, or even just e, ou > o/e. I'm not sure on what all vowels change into but those are what I have found so far.

G is also prone to elision, e.g. 如果 > ruo

q, j, x, also become y/i in the North or can become c, z, s, in the South


This doesn't pertain specifically to Northern Mandarin, but many of these contractions are found outside of Taiwan in this article https://chinesehacks.com/vocabulary/syllable-contractions/


Here are some Cantonese contractions. denoted with Jyutping (J) and IPA (I)

廿 (uncommon: 廾) J: /jaa6/ I: /ja22/, contraction of 二十

卅 J: /saa1/ I: /sa55/, contraction of 三十

卌 (uncommon character) J: /se3/ I: /sɛ33/, contraction of 四十

However, 三十, 四十 are usually read as a slurred contraction of 2 characters instead of the special contracted character.

三十 J: /saam1 sap6/ I: /sam55.sɐp2/ may be contracted as J: /saa1aa6/ I: /sa52/

四十 J: /sei3 sap6/ I: /sei33.sɐp2/ may be contracted as J: /sei3aa6/ I: /sei33a2/

五十 J: /ng5 sap6/ I: /ŋ̩13.sɐp2/ may be contracted as J: /ng5aa6/ I: /ŋ̩13a2/ or with "lazy sound" I: /m̩13a2/


  • shouldn't the tone for 卅 be 53, not 55? Commented Feb 18, 2021 at 0:31
  • @kwaalaateimaa technically speaking, the tone 53 does not fit in the Hong Kong standard of 9 tones. However, broadly speaking, the first tone (陰平 high level tone) may be pronounced as either 53 or 55. Older generations may still differentiate 53 and 55 and consider them different words. However, on a personal note, I believe 卅 is usually pronounced as a slurred 三十, causing the tone to gradually change, also noticeable in the length of slur sounding more like 2 characters are being spoken as well as the last tone is more like a 2 than a 3. Commented Feb 18, 2021 at 3:16
  • ... comparing 2 (下 /ha22/) 沙下 (saa1haa6) and 3 (啊 /a33/) 沙啊 (saa1aa3) Commented Feb 18, 2021 at 3:22
  • some words are almost always pronounced with the 高降 tone, not 高平, such as 咩,添,先,親。Maybe 55 and 53 are largely allotonic, but are distinct in some cases. If you use the high level tone for 先 for example, it sounds like you're talking down to someone. Commented Feb 18, 2021 at 4:57
  • @kwaalaateimaa I do wholeheartedly agree, that's why I added my second half of my previous comment comparing the tone of 卅: 沙下 VS 沙啊. And according to Wikipedia Cantonese Phonology "...sàam (high falling) means the number three 三, whereas sāam (high level) means shirt 衫". IMHO, the tonal gradient of this high falling 53 is different than what is usually spoken with 卅 52. Perhaps the other answer might be there are different gradients yet explained by linguists, (or most likely I'm ignorant of these researches). Commented Feb 18, 2021 at 6:14

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.