Mandarin, as everyone probably knows, has four tones: 1st, 2nd, 3rd & 4th tone.

Each tone has its own tone mark that represents that tone:

  • 1st tone = ˉ (macron)
  • 2nd tone = ˊ (acute accent)
  • 3rd tone = ˇ (caron/háček)
  • 4th tone = ˋ (grave accent)

But these are all just representations of each tones actual 调值 (tone value), i.e.:

  • 1st tone = 55
  • 2nd tone = 35
  • 3rd tone = 214
  • 4th tone = 51

Each tone mark is a visual representation of each tones value. Good? Good.

Now the question: Do any dialects have tone marks?

Cantonese, at least the works that I have seen, is written with pinyin and tone numbers, i.e: gwong2 dung1 waa2...

Other topolects are simply written with IPA + tone contour marks.

  • 2
    The Yale romanization of Cantonese uses tone marks (and -h for the lower series). There is a wiki page about this.
    – neubau
    Commented Jul 1, 2014 at 5:07
  • Had a look. Isn't 2nd tone and 5th tone's tone mark the same? (A rising acute accent mark?) How would one differentiate between them?
    – Mou某
    Commented Jul 1, 2014 at 5:11
  • 2
    Tones 4,5,6 and 9 (the low ones) have an -h after the vowel, but before any final consonants.
    – neubau
    Commented Jul 1, 2014 at 5:16
  • Do you strictly mean topolects (e.g. Cantonese, Wu), dialects (e.g. Shenyang dialect, Beijing dialect) or both? Commented Jul 1, 2014 at 7:03
  • topolect is just a proper translation for 方言. I wrote both to cover all the bases. To answer your question both.
    – Mou某
    Commented Jul 1, 2014 at 7:08

3 Answers 3


You could look at Tianjin, as close as 120 km from Beijing. The local Tianjin dialect is still exhibiting rather drastic changes to tones and tone changes.

Drawing from your example, mandarin in Beijing and Tianjin differ in tones according to the following, with TJ accent being much more low-pitched:

  • 1st tone = 55 (BJ), 21 (TJ)
  • 2nd tone = 35
  • 3rd tone = 214 (BJ), 113 (TJ)
  • 4th tone = 51 (BJ), 53 (TJ)

You also have different sandhi rules for tonal changes. Tianjin itself is pronounced Tian3jin1 in TJ. Tone marks, however, would still be identical (just as "pen" is not rendered "pin" in Texas).

This pattern repeats itself wherever you go in China. For example, the standard Wu topolect as spoken in Shanghai has differences already within the city suburbs, such as Jiading, and much more so in neighboring cities like Jiaxing, and further more in relatively remote locations such as Hangzhou and Suzhou.

From this we conclude that the standard mandarin (普通话) is merely an ideal, a common framework for speakers of countless topolects and dialects.

  • "Tone marks, however, would still be identical (just as "pen" is not rendered "pin" in Texas)." Sure but Mandarin tone marks are representative of the tones themselves - flat is flat - rising rises, etc etc. Wouldn't it be more intuitive if dialects tone marks also represented their tones?!
    – Mou某
    Commented Jul 11, 2014 at 14:13
  • It would lead to a zoo of such tone marks. We are better off using tone numbers, and in actuality only linguists would have use for them. Learning Tianjin accent knowing standard Mandarin is like learning Nu Yawk accent knowing some standard English pronunciation.
    – user4452
    Commented Jul 11, 2014 at 14:35
  • Sure but the whole point of my question was tone marks, right? I mean it's in the title, no? I think tone marks would be more useful for non-linguists rather than linguists, I mean you look at Mandarin it's flat, rising, falling-rising and falling - that's nice and all but completely lacking precision - a linguist would be more interested in tone values if anything.
    – Mou某
    Commented Jul 11, 2014 at 15:27

Judging by the word topolect, I think you are most interested in this: Phonetic realization of Mandarin tones in principal dialects (scroll down to the table with the caption 'Phonetic realization…'). This is how people from different regions will most likely realize the tones when they are speaking mandarin with the "local flavour", whereas educated people, tv/radio hosts will perhaps strictly stick with the Standard Mandarin, even if they are not from Beijing.

Dialects have their own tone systems, with a varying number of tones. Unfortunately, phonological material is only available for a very few dialects, e.g. Hokkien, Cantonese, Shanghaiese [if you consider these to be dialects, I think they are languages on their own]. I used to do research on the tone systems of tonal languages, and there are at least two types of notations, the one with numbers describe the pitch, just as the OP demonstrates, and there is another kind of notation which uses "sticks", like ˥ Mandarin 1st tone, etc. (unfortunately this site doesn't support pasting them, please refer to the table in this article.

The Chinese people, when they are to discuss the tones for any reason, including linguists, will use tone names, that have existed since centuries if not millennia, e.g. 阳平、去声、上声, etc. These are less accurate, since they are not absolute notations of a tone contour and could vary across dialects and regions. A good reference would be this table.


You might be interested to know that for some of the minority (non-sinitic) languages of China with tonal systems, tone letters are used rather than marks or numbers. Two examples are Zhuang and Hmong (Miao). Some examples:

Northern Zhuang

na [na24] ‘thick’

naz [na42] ‘paddy field’

naj [na55] ‘face’

So, no letter indicates a rising tone, -z a falling tone, -j a high tone. There are 6 open syllable and 2 closed syllable tones altogether. (All of these words have clear cognates in Thai, by the way.)

White Hmong

kuv [ku24] ‘I’

koj [ko53] ‘you’

pos [po11] ‘thorn’

How confusing this would be depends in part on whether real consonants can appear in syllable-final position or not. If so, you would have to get used to distinguishing the real ones from the ones just marking tone. For these two languages, I believe Zhuang has a fair number of final consonants (including stops), but Hmong has very few.

I don’t know if this system has ever been applied to a Chinese dialect. Probably one could do so, because Chinese syllables are not that different from Zhuang and Hmong ones in their basic shape.

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.