10

I'm just reading a book about China and Chinese customs from Heike Barai. She says that Chinese only has around 300 different syllables, whereas modern German has about 10 000 and I suppose English is about the same.

Does Chinese really only have 300 syllables?

2

3 Answers 3

22

I took the CEDICT file and wrote a script on it. The file has 113k dictionary entries, so it covers a very large portion of the Chinese vocabulary.

There are 1522 different pinyin syllables in CEDICT, when you distinguish tone numbers (like, ma1, ma2, ma3, ma4, ma).

If you do not care about tone numbers, you'll get 413 syllables (ma, mo, mi, etc.)

These results include m2 and m4 (呒, 呣 [interjections]).

UPDATE: Of course, you can argue whether 'ging' and 'fiao' are Mandarin pinyin syllables or not, but they are in the CEDICT file and my answered is based on that.

9
  • 1
    Syllables such as fiao and ging are simply not Mandarin. I wonder where CEDICT got these?
    – wpt
    Aug 9, 2015 at 14:20
  • 4
    @wpt 覅 fiào is a contraction of 勿 and 要, so this is colloquial. Ging has no character, it is a Minnan loanword, in zhuyin fuhao it is ㄍㄧㄥ en.wiktionary.org/wiki/%E3%84%8D%E3%84%A7%E3%84%A5 "ㄍㄧㄥ" returns more than 300k results on Google, almost exclusively used by Taiwanese.
    – imrek
    Aug 9, 2015 at 14:30
  • 1
    Where do you guys get all this info? Wish I could use the Internet like that! Nonetheless, all praise for the Drunken Master! A master work. I'll try to contact Frau Barai and tell her。 Could you send me that script? I'm running Ubuntu 14.04. I'd love to see it!
    – Pedroski
    Aug 10, 2015 at 22:52
  • 1
    Sorry, I have accidentally already thrown out the script, if you can wait until today evening, I'll try to rewrite and post it a link to it. You will need to run Python. I'm going to write you some instructions, if you have no experience with it.
    – imrek
    Aug 11, 2015 at 6:55
  • 2
    @Pedroski I have rewrote the script and uploaded to pastie: pastie.org/10344191 Here's how you can run it 1) create folder on your system and 2) copy the full script from pastie into a new text file (use gedit on Ubuntu) and save it as pinyincount.py (the extension is important). 3) Download the CEDICT file from here mdbg.net/chindict/export/cedict/cedict_1_0_ts_utf-8_mdbg.txt.gz 4) Extract it to the same folder as the script. 5) Open a terminal window, and navigate with 'cd' to the folder in which the script and the cedict file is. Run with python pinyincount.py
    – imrek
    Aug 11, 2015 at 15:58
11

As Maroon points out in a comment, you have to say which dialect you are asking about. This answer is for Standard Chinese, aka Mandarin.

It also depends on what sort of stuff you include. Counting the distinct lines in the syllable index of the Pinyin Chinese-English Dictionary, I get 420 "lines" but this includes some very marginal stuff such as tei, kei, den, etc. This does not include any of the rhotacized syllables, where a suffix -er merges with the syllables and produces many distinctive syllables.

Most important, however, this figure ignores the four lexical tones of Mandarin, which is definitely not the right; tones are distinctive in Mandarin, so syllables with different tones are different syllables. The distribution of tones and syllables is skewed, however; not every syllable has all four tones. A rough approximation is about 1200 distinct syllables. Again, there will be many marginal cases.

All this does not use any sort of phonemic analysis which could perhaps reduce the number of syllables, depending on what phonological theories or notation you adopt.

So Chinese is not so syllable poor as you might expect. Japanese and Hawaiian, for example, have far fewer distinct syllables than Mandarin.

1
  • Oops, I deleted that comment after writing my answer, since I basically included the idea there anyway. But anyways, +1 for making mention of a number of the things I missed and generally dealing well with the point.
    – user5714
    Aug 9, 2015 at 6:11
6

The answer to this could well depend on the specific language meant by "Chinese". For instance, Cantonese supposedly has over 600, which isn't surprising because of larger variation of vowels and the presence of more ending consonants. (I am unsure as to whether this figure takes tones into account.) I will assume that this is about 普通话, since that seems to be generally taken as a default.

普通话 is easier to assess, because pinyin charts with all possible pinyin combinations are easily available, and generally a syllable simply corresponds to a distinct combination pinyin. (I am not sure if different tones count as distinct syllable combinations -- the estimates would appear to be really off if they were, so I am not going to take those into account.) Thus, I copied this table into Microsoft Excel and deleted all syllables that were either duplicated or that do not, in my memory (from formal Mandarin lessons), appear in 普通话, with some cross-checking by typing pinyin combinations I was suspicious about into this online dictionary.

The result I'm currently getting is 408. This is slightly higher than the number your book reports, but in line with Wikipedia's. So 300, at least for Mandarin (which would make the most sense as an object of discussion), is probably a slight underestimate, particularly if the writer is thinking of something towards the lower end of the 300's.

3
  • 1
    Fast work! I wonder if there are better references than you or I have, though, something more formal.
    – wpt
    Aug 9, 2015 at 6:44
  • 3
    Cantonese has 9 tones. If you took it into account... it's simply impossible! The 9 tones can be represented in the sound si: 詩史試時市事色錫蝕 or simply 三九四零五二七八六(394052786). Aug 10, 2015 at 10:03
  • 408 is not correct for Mandarin either. More than 1200.
    – 伟思礼
    Jan 18, 2018 at 8:13

Your Answer

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

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