Jump to content
  • Sign Up

Cantonese tones 2/5 and 4/6 confusion


Recommended Posts

I'm Teochew but am also fluent in cantonese due to my upbringing. I never knew there were 6 tones in Canto (minus the -p -t -k endings which IMO are BS tones because it's really the same thing as 1 3 6 as people have said) until I did some searching online and also when I sorta learned the romanization systems so that I can look up words when needed.

However, I seriously cannot see the difference between 2/5 and 4/6. Well, 2/5 I can sorta tell, but 4/6 sound identical to me!

時 (4) vs 事 (6) sound exactly the same to me

史 (2) vs 市 (5) sound identical to me too (albeit I can sorta tell a slight difference but it doesnt seem like a big deal to me, as in I can use 2/5 interchangeably and people can understand me with no problems at all).

One way to memorize the tones is through the numbers 394052 for the 6 tones. Well, as mentioned '9' and '5' are 2/5 respectively but sound the same. '0' and '2' are 4/6 but sound EVEN MORE the same.

I asked some of my friends who are native canto speakers and they don't differentiate between the 2 tones either. They tell me when they say those character pairs up above, each pair has the same tone.

My question is, is it such a big deal? tones 4/6 and 2/5 can pretty much be interchanged and it sounds pretty much the same. In normal conversations I can't see how anyone can tell when someone mixes up 2/5 or 4/6?? As I mentioned, 2/5 MAYBE (but really anal I'll say) , but 4/6 hell no.

IMO anyone who calls someone out for mixing 2/5 or 4/6 is simply being anal. I don't mean to offend anyone but really, cantonese can pretty much be spoken with the 5 mandarin tones (including the neutral tone). Even in my first language teochew, which apparently has 8 tones, I can pretty much think through all the different words I know and match each tone with one in mandarin.

Are the subtle differences (though i can't tell im sorry) that big a deal? In a normal conversation you wouldn't even be able to tell right?

Again, no offense, but I just want to know what you guys think, or at least enlighten me if there's something I'm not understanding?

PS. I know people can probably come up with words that have totallly opposite and inappropriate meanings when you switch up tones 2/5, but really, this proves nothing, as in a normal conversation + context, nobody is gonna notice anything if you happen to say a word that means iono "food" in tone 2, but you say tone 5 instead and it sounds like a word that means "boobs". Whatever difference if any is too subtle, plus with context nobody is gonna think you mean "boobs".

Edited by MXQBLGH
Link to comment
Share on other sites

Having just spoke to my friend who is a BBC, she is unable to tell the difference between 4 and 6, and only just between 2 and 5, so it seems you are not alone. I hope more people can give some input as I think this is a very interesting discussion.

Link to comment
Share on other sites

I think it is difficult to distinguish these tone pairs isolation. But in words or expressions where both tones appear next to each other difference is a bit more obvious. For example:

飲嘢 (2 5)

女仔 (5 2)

豉油 (6 4)

問題 (6 4)

承認 (4 6)

Edited by sebhk
Link to comment
Share on other sites

thanks for the examples

飲嘢 (2 5)

女仔 (5 2)

I can tell the difference only slightly, iono but I think it's being "forced" just because you've given me which one is 2 which is 5. The first pair is 25, second is 52, but lol I probably wouldnt know that unless you told me. Iono, to me it seems more like, in my natural speech, the "yeh" in "yum yeh" is a bit 'shorter' in pronunciation. Similarly with "nui" in "nui zai", the nui is shorter, while the zai is a bit longer thus making the differencein low rising vs high rising. I guess if I hadn't known the tones, I'd just think this difference is just a natural part of speech, with more emphasis on certain words.

I guess what I'm saying is if I say each pair of words in tones 5 only, or tones 2 only, or mix up the wrong combo of 25 or 52, it wouldn't matter because in normal speech it sounds close enough, if not the same.

豉油 (6 4)

問題 (6 4)

承認 (4 6)

Each of these pairs sound identical to me lol.

Again, whether I say 66, 44, 46, 64, it sounds the same to me and I don't think anyone could possibly misunderstand me. Perhaps I'm just not producing the tones correctly, but hey, I'm using different words as examples, or I'm using the numbers 0 and 2 which sound identical to me.

This is interesting to me because it seems like some people just like to brag that their language has more tones and is thus harder to learn. When in fact, these number of tones (greater than the 5 distinct mandarin tones) really only have additional tones that sound pretty much the same (high rising vs low rising for instance), or are restricted through endings like -p -t -k, which IMO are pure BS tones. The number 7 in cantonese should not be a tone 7, it should simply be a tone 1, for instance. Why make it so difficult and complicated lol.

I speak teochew (chaozhou) which has 8 tones, but really I can use the 5 mandarin tones to match every single word I can think of in the language. The additional 3 are prolly something like a "-k -p -t" ending restriction or a "high rising vs low rising" tone difference, both which are subtle and make no difference.

Iono, I just think that mandarin is a perfect language to compare with because it really has 5 distinct tones which are easy to tell apart (by neutral tone in mandarin I mean the word like '的' which is pretty much equivalent to cantonese tone 3).

Link to comment
Share on other sites

Even for Mandarin, all tones are not equal! If you take a phonetics course in China, you will soon learn that Mandarin has more than just four tones. Its second, third and fourth tones all undergo slight changes (usually referred to as half-tones) in different combinations of tones. One could just as easily say that for Mandarin, there are more than four tones + a neutral. Mandarin half tones could just as easily be classed as tones. Certainly, if you want to sound like a more natural speaker, you cannot treat all tones as fixed. As for the Cantonese given, you need to be certain that the person saying it is a standard Cantonese speaker (and what that standard is, Hong Kong Cantonese, Guangzhou Cantonese, overseas Cantonese...). There are lots of variants of Cantonese, for some, the expressions given do not highlight any tonal differences, but for others, the difference is discernible if your ear is attuned to it. Does it make a huge difference at the end of the day, probably not.

Link to comment
Share on other sites

Interesting thread.

The native speakers I know (Malaysian and Hong Kongers) would never confuse tones 2 and 5 - the pitch difference is too distinct. If I ever mixed up tones 2 and 5 around my native-speaker Malaysian boyfriend, I would definitely get corrected, no doubt about it.

However, I can get away with pronouncing tone 5 as tone 3, probably because, as the OP has noticed, tone 5 is somewhat 'shorter' and ends at the same pitch as tone 3.

As for tones 4 and 6, I could probably get away with mixing those two up (at least compared to 2/5) but native speakers can hear the difference. I think that tone 6 is 'shorter' than tone 4, like the way tone 5 is 'shorter' than tone 3.

I agree with the OP that it's pretty unlikely you'll ever be misunderstood for mixing up 4/6 and 2/5. I think it's more a case of sounding like a 'native' as opposed to someone who is just 'fluent' (eg native speakers of other Chinese dialects who live in Cantonese-speaking communities, such as some native Mandarin speakers in the HK entertainment industry).

That said, I have read on one Cantonese forum that native Cantonese speakers in Macau actually do pronounce tones 2 and 5 the same way. So maybe it's not quite that big a deal :D

Link to comment
Share on other sites

Kk i found this vid on youtube which was really helpful in differentiating the tones


Yeah I guess I can sorta hear the difference now that I really listen closely.

Lol, but I still stand by the fact that you can pretty much substitue 2/5 with mandarin 2 and 4/6 with mandarin neutral (for particles like 'de') and you'll be understood.

I don't know, I just find that during a regular conversation, nobody (well except natives I guess) speak so perfectly as to get the tones exactly correct. There's room for error and those slight discrepancies aren't too noticeable in normal speech. I can say '女仔' in a normal sentence as 2/2 or 5/5 and it wouldn't even be a noticed. Seriously who listens for this stuff. The two tones are close enough to be the same.

I'm not saying there's no difference, I'm just saying I, like many of my friends, grew up with the language (albeit in Canada, not HK), and didn't even know there was a difference in the two tones for the words 女仔, or the other examples I listed above. si2 vs si5 for instance.

Link to comment
Share on other sites

As for tones 7,8,9. They may be different, but I think it doesn't add anything but confusion to someone learning the tones. Why have baat8 (for eight) just because of the -t ending when baat3 would accomplish the same thing. All they're doing is classifying -t -p -k endings as tones themselves when they are simply 1,3,6 with those endings. If we're talking tones/pitch, then how the word ends shouldn't matter right.

Link to comment
Share on other sites

Seriously who listens for this stuff.

It's not that people listen - consciously listen out - for this "stuff", as you put it. It's just that, to a native speaker, it sounds "off" and unless they're used to hearing people (eg other non-native speakers) speak that way, they will notice it.

It's like when someone speaks English with a non-native accent and puts the stress on the wrong syllables, it doesn't (often) result in them being misunderstood, but it sounds "off". (Not meaning to pick on non-native English speakers, it's just an example that I encounter quite frequently :) )

Whether you think the difference is substantial or not is a matter of personal opinion. I also think that non-native speakers tend to have trouble differentiating between certain differences in sound which do not exist in their native language. You might think that Mandarin tones are very distinct, but when I demonstrated them to my non-Chinese, English-speaking French class, they couldn't tell the difference at all :wink:

Link to comment
Share on other sites

Haha, ok I might've sounded like a douche and I apologize for that if I came across as one. I just want to note that I do acknowledge that there IS a difference between 2/5 and 4/6. Otherwise, if there wasn't, then it would be one hell of a coincidence that all these websites, and books, all say the same thing ::mrgreen:

But yeah, anyway, I asked some more native canto speakers and they noted that there was a difference between 史 and 市. I didn't bother asking about 4/6 though.

Yeah and you make a good point, that unless the native speaker is used to hearing foreigners speak with the wrong tones, it will sound off.

I guess I should try my best to correct my 2/5 and 4/6 tones.

I still stand by what I said about tones 789 though.

Link to comment
Share on other sites

By the way, here's some information posted on another forum about the various Cantonese dialects and tone mergers:

According to 廣東粵方言概要 (An Outline of Yue Dialects in Guangdong published in Guangzhou by Jinan University Press, 2002.7) written by Professor 詹伯慧, the Yue language (粵語) consists of 4 major dialects as follows:

1. Cantonese (廣州話) – Major cities include Guangzhou, 番禺, 增城, 佛山, 南海, 三水, 順德, Shenzhen, Hong Kong. This is the standard dialect in Guangdong and the only one that consists of all the 6 basic tones. However, 從化 has tones 2 & 5 merged. Macau, affected by 珠海 on its north, has lost tone 2.

2. Taishanese (台山話) – Major cities include 台山, 恩平, 開平, 新會, (珠海)斗門, 江門, 鶴山 (part). This dialect has only 5 tones, with tones 3 & 6 merged. But 恩平 has no tone 5, i.e. it has a total of only 4 tones.

3. Shiqi (石歧話, also known as 香山話): Major cities include 中山, 珠海 (except 斗門). This dialect has only 4 tones, with 2 & 5 merged and also 3 & 6 merged.

4. Dongguanese (東莞話) – Major cities include 東莞, 深圳寶安, 新界圍頭話. Tonal variations are great for this dialect. For instance, 莞城 has tones 3 & 6 merged, 沙井 tones 2 & 5 merged, 錦田 tones 3 & 6 and also 2 & 5 merged.

What I am trying to say is that most speakers of non-Cantonese dialects have no tone 2 or 5 in their speech. I am certain that many of them have emigrated to Hong Kong, Southeast Asia, North America, etc and their children still retain their dialectal variations.

(The original post can be read here and the thread is here.)

Perhaps this might explain your confusion.

Link to comment
Share on other sites

One thing that has helped me is listening to CD materials (with the book and corresponding Yale Romanization and tones). Then you can listen to one section over and over, and specifically do drills to listen for one partcular tone that is giving you trouble.

Or, another good book, is Fun with Cantonese: Sounds and Tones. There are specific chapters on 2 vs. 5 and other common confusions.

Link to comment
Share on other sites

I still stand by what I said about tones 789 though.
Tones aren't distributed at random. There are many Chinese fangyan, and not only their sounds are related, their tones are too. I don't know any Cantonese, but for Minnan, first tone Minnanese is almost always first tone Mandarin; third tone Minnanese is almost always fourth tone Mandarin; and if you're a linguist, you can make the list much longer and more comprehensive. I would guess that something similar is going on in Cantonese, and if you just start considering two different tones the same, you mess up these relations.

If you don't need those distinctions to make yourself understood, by all means, ignore them. But don't call the people who do see a difference stupid, they have very good reasons for their ideas.

Link to comment
Share on other sites

  • 1 month later...


While tones 2 and 5 are rising tones, they're distinctive even when considered in isolation because tone 2 rises about a major 3rd (2 whole chromatic tones), while tone 5 rises a major 2nd (a whole tone). If you have a musical ear, you can tell the difference. There's no difference in length — that depends only on the length of the vowel/diphthong, not the tone.

Tones 4 and 6 are tricky. Many Cantonese speakers pronounce both as level tones, i.e. no rise or fall, so it's impossible to tell between the two if they're pronounced in isolation. 4 is simply sounded at a lower pitch, usually a semitone lower than 6.

On the other hand, some Cantonese like myself pronounce tone 4 as a falling tone. In this case a tone 4 word like 華 or 羅 sounds distinctive to the ear because of the falling pitch, even when said in isolation.

Whether pronounced as a level or falling tone, tone 4 has the lowest pitch of all the 6 (or 9) tones in Cantonese.

Correspondence with Mandarin tones (general rules):

1. Cantonese 1 = Mandarin 1

2. Cantonese 2 = Mandarin 3

3. Cantonese 3 = Mandarin 4

4. Cantonese 4 = Mandarin 2

5. Cantonese 5 = Mandarin 3 & 4

6. Cantonese 6 = Mandarin 4

Edited by Yan Hoi
Link to comment
Share on other sites

@Lu and others

just about tones 789, which are from 入聲 in Middle Chinese. This category is notorious in historical phonology. While the other tones, 陰平,陽平,上,去 have relatively straightforward correspondeces in Modern Mandarin, 入聲 are distributed across all four tones in Modern Mandarin. These might still turn out to be regular corrspondences of extremely complicated sound changes (I don't know since I'm not a historical phonologist) rather than random, but I would be surprised if tones in Mandarin and Cantonese that were 入聲 in Middle Chinese would correspond to each other in a straightforward manner.

Link to comment
Share on other sites

I would be surprised if tones in Mandarin and Cantonese that were 入聲 in Middle Chinese would correspond to each other in a straightforward manner.

入聲 in Mandarin seems randomly distributed, but homophones in Cantonese usually map to homophones in Mandarin. One could say that 入聲 works only on a syllable-by-syllable basis. For example,

識 and its homophones are usually shik7 in (nonstandard) Cantonese, and shi4 in Mandarin.

錫, sik7, xi1

殺, shaat8, sha1

石, shek9 shi2

剔, tik7, ti4

You may want to see

video. on Middle Chinese to Modern Chinese mapping.
Link to comment
Share on other sites

great link thanks. So we can say as follows:

- 入聲 with 全濁 (voiced obstruents) initials in Old Chinese becomes 2nd tone in Mandarin

- 入聲 with 次濁 (voiced nasals and approximants) initials in Old Chinese becomes 4th tone in Mandarin

- 入聲 with 清 (voiceless obstruents) initials in Old Chinese have unpredictable tone in Mandarin.

- Shanghainese and Taiwanese split their 入聲 neatly into two sets depending on whether they were voiced or voiceless in Old Chinese: a neat correspondence: 清入 in Old Chinese becomes 陰入 in the modern varieties and 濁入 in Old Chinese becomes 陽入 respectively.

- Cantonese also follows the same split along the lines of 清濁, however it also splits its 陰入 according to the vowel length, e.g. a higher tone for a short vowel and a lower tone for a long vowel.

The table below is a simplified version of one found in the Wikipedia:


So we can see that for 濁入 we have a neat correspondence between Mandarin and the other three here. Taken from the same link:

  1. 全清入:一八百法黑國德北節血必雪骨竹室色發識屋
  2. 次清入:七鐵出塔尺踏赤冊乞泣撲哭
  3. 全濁入:十白學直絕獨石熟實蝶達
  4. 次濁入:六入木葉日綠月落莫滅立密麥肉玉

3. in Mandarin should all be 2nd tone, 4. in Mandarin all be 4th tone, and 3. and 4. should be 陽入 (9/6) in Cantonese.

1. and 2. in Mandarin are all over the place, however in Cantonese they should be 陰入 split up according to whether the vowel is long (8/3) or whether it is short (7/1).


a. what does the vowel length distinction in Cantonese correspond to in Old Chinese or Middle Chinese?

b. If you pronounce the Mandarin and Cantonese tones in the groups 1. and 2., are there any correlates at all, or is it totally random? (Now what Hofmann says about all homophones being equal makes sense since sound change usually tends to affect sounds uniformly. The seeming randomness of 入聲 could actually be based on quite regular sound correspondences as well, it could just be that they are not well understood yet)

Link to comment
Share on other sites

what does the vowel length distinction in Cantonese correspond to in Old Chinese or Middle Chinese?

From my perspective, vowel length distinction was something that developed after Middle Chinese. One could just ignore the long/short vowel concept with Cantonese and just recognize different vowels as different vowels. Individually, they correspond to different vowels in Middle Chinese, but I haven't seen any vowel length distinction in Middle Chinese. Therefore, a connection between vowel length in Cantonese and something in Old Chinese is improbable.

If you pronounce the Mandarin and Cantonese tones in the groups 1. and 2., are there any correlates at all







(踏 is weird in Cantonese.)

It appears that it's random.

Link to comment
Share on other sites

Join the conversation

You can post now and select your username and password later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Click here to reply. Select text to quote.

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...