I'm reading prosodic phonology, and wondering if there is any re-syllabification process happening in Mandarin Chinese?

  • I do not know Mandarin at all. However, I believe I've heard of possible contractions formed from multi-syllable words that smush together different syllables. Would that sort of thing count? Feb 26, 2016 at 22:20
  • By the way, what is the definition of "resyllabification"? I've been trying to find it. Feb 26, 2016 at 22:56
  • 1
    The resyllabification here means the processes like French Liaison rules.
    – chaoh
    Feb 28, 2016 at 4:09

3 Answers 3


(note: this answer is not complete, I am not a linguist and I have no personal knowledge of Chinese. I hope it is still of some use despite this.)

I have found several sources that say that resyllabification does not occur in Mandarin Chinese. However, it is possible that they are just making general statements, and not mentioning some rare examples.

syllables in Chinese are not re-syllabified in connected speech, for instance, 西安 (/xi1.an1/, capital of Shanxi Province in China) (a dot indicates a syllable boundary and the number indicates tone) cannot be falsely identified or pronounced as/xian1/(e.g., 鲜, fresh). In contrast, the syllable boundaries of a phonological word in Dutch or English differ from a lexical word’s canonical syllabification. For instance, for the word “predict” (pre.dict), different contexts require a different re-syllabification, such as “predicted” (pre.dic.ted) when using the past tense and “predict it” (pre.dic.tit) when producing the cliticization.

(Masked Syllable Priming Effects in Word and Picture Naming in Chinese, by Wenping You, Qingfang Zhang, Rinus G. Verdonschot)

Mandarin has a limited inventory of syllables; syllables have a very simple structure and re-syllabification is absent.

(The Proximate Phonological Unit of Chinese-English Bilinguals: Proficiency Matters, by Rinus Gerardus Verdonschot, Mariko Nakayama, Qingfang Zhang, Katsuo Tamaoka, and Niels Olaf Schiller)

However, I also found this:

Yin (1986) proposes that the retroflexed rime in Beijing Mandarin is the result of re-syllabification of the stem and the retroflex suffix.

Mandarin Retroflex Suffixation: An OT Account, by Ma Qiuwu

The referenced paper is cited as follows:

Yin, Yun-Mei. 1986. An Autosegmental Approach to Retroflex Suffixation and Reduplication in Chinese. Taipei: The Crane Publishing Co.

Unfortunately, I haven't been able to access this book yet, but I may be able to request it from my library. It seems in addition to the retroflex suffix, it also discusses reduplication.


Cholin 2010 ("Do syllables exist"), Chen, Chen & Dell (2002) ("Word-form encoding in Mandarin Chinese as assessed by the implicit priming task") claim there is not. Duanmu (1992: dissertation) claims that onsets are obligatory, so C#V resyllabification would not arise. However, he notes p. 18 that there are "weak interjection syllables" which allow C-V linking, viz. tian-a "Heavens!". This is consistent with the view that there is no resyllabification, as long as you don't initially syllabify this as .tian. + .a. which he accomplishes by claiming that the domain of syllabification is the word plus any weak suffix.


This assumes what has become the standard theory of what resyllabification means: where a syllable is formed, then a morpheme is added or segments are deleted or inserted, and then syllable structure is re-assigned. Since there is no relevant epenthesis or deletion in Mandarin, only concatenation of words would count.

  • For syllables that seem to start with "a," is the obligatory onset considered to be a glottal stop? Feb 26, 2016 at 22:55
  • Duanmu allows glottal stop plus γ (not a phoneme) and geminate [ŋ], so /mian ao/ "cotton coat" → mian ʔao, mian γao, miaŋ γao and miaŋ ŋao (and *mianao)).
    – user6726
    Feb 26, 2016 at 23:08

Would you count erhua as a form of resyllabification? The er invades syllables like a facehugger to people.

  • 哪儿 “where” (érr /naɚ̯˨˩˦/)
  • 一点儿 “a little” (yì diǎn éryì diǎnr /i˥˩.djaɚ̯˨˩˦/)
  • 好玩儿 “fun” (hǎo wán érhǎo wánr /xau̯˨˩˦.waɚ̯˧˥/)
  • 一瓶儿 “one bottle” (yì píng éryì píngr /i˥˩.pʰjɚ̃˧˥/)

In the last example, the ér literally replaces the nucleus of the prior syllable leaving only the initial and the medial intact.

  • I don't think this really counts, because -r and -ér aren't the same morpheme; the former is a suffix mainly with diminutive meaning, and the latter means 'son'/'child', and so there's no reason to think of -r and -ér as allomorphs synchronically. Jun 25, 2017 at 2:05

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.