The hardest tone in Chinese

[Shi Tong]

It also depends on the position of the third tone within the word and whether the word is stressed or not. Listening in on your wife making a phone call is a bit too little for such a blanket statement (also if you did record her call look at the graph of her fundamental frequency whether or not there is a small less perceptible increase towards the end).

Chrix, remember that I've done listening to my wife on the phone, talking to friends, spending time with family, in Taiwan, with other Taiwanese people etc etc etc for 9 years so far.

What I meant was that I was very careful to listen out for any rises at the ends of 3rd tones at the ends of sentances, I was very careful to take note of the ones in the middles etc.

Take a look at

, I found on youtube under Taiwanese news.. note that at 19 seconds in, she says "shuo3" for hand, and this is most definately a flat low tone, and then at 34 seconds, I would say she mispronounces "shi2" (time) and says it as a low tone also.

At 59 seconds, there is a "zu3", but I'm not sure if this is qingsheng or not, but if it's supposed to be a 3rd, then it's low and flat. Again at 1:33 there is a "shi3" at the end of the sentance there and, again, low and flat.

Am I just going mad or are these actually qingsheng sounds which I am assuming are 3rd tones.. they all sound low and flat to me, and this is the NEWS...:lol:

Dont get me wrong, Chrix, I know it's taught as a falling and rising tone, but I always experience it as a low flat tone..

[renzhe]

If your wife cooks for you and asks you "好吃吗?", and you answer "很好!", or if she asks you if you want to go to the zoo "我们去动物园吧!" and you answer "好!", how do you pronounce 好?

A flat low tone? Pronounce the falling part but not the rising part? Really?

The third tone is very flexible, and I find that it is affected by sandhi and sentence-level prosody more than other tones, but this is the first time I hear that there are people who NEVER pronounce the tone fully, even in isolation. I'm even more puzzled that there are people who have been learning for 9 years and have never heard a native speaker use it in speech.

[creamyhorror]

The rising tone at the end of the 3rd tone in isolation makes it sound "proper" (or foreign) to me, like you might see on a programme from Northern China. Guess it's just the lousy pronunciation here, though.

I'll try listening to a few clips from local news broadcasts tomorrow and see if they pronounce the rising section on 3rd-tone endings.

[renzhe]

It's hard to make general statements on tones based on rapid-fire 200 wpm news snippets because tones come so fast that they tend to blend into each other. The newscaster Shi Tong posted has such a staccato accent that there's no time to pronounce the full 3rd tone even if she tried.

But here's one for you:

Pay attention at second 4 (当晚), 10 (背角 I think), 19 (三角), 23, 25... Edited April 8, 2010 at 11:04 PM by renzhe

[wushijiao]

But here's one for you:

Pay attention at second 4 (当晚), 10 (背角 I think), 19 (三角), 23, 25...

I think that clip certainly shows the difference in the stressed vs. unstressed tones (especially the third), as I mentioned in one of the earlier posts, going off this info about the crazy nature of third tones.

I think Shi Tong's notion that the 3rd tone might stay consistently low in Taiwanese Mandarin is interesting. (I would assume there be piles of academic literature on the topic, no?).

[creamyhorror]

But here's one for you:

Pay attention at second 4 (当晚), 10 (背角 I think), 19 (三角), 23, 25...

04: I hear it

10, 19, 23: hear it, not a very strong rise though

25: (发光体) doesn't seem to have much of a rise

I guess we just don't speak like that in Singapore. Here's a clip of Singapore Mandarin (spoken by the two hosts):

http://video.xin.msn.com/watch/video/new-city-beat-casual-shoes-%E5%9F%8E%E4%BA%BA%E6%96%B0%E6%9D%82%E5%BF%97-%E4%BC%91%E9%97%B2%E9%9E%8B/tn2ha59i

(1.27 非常适合你)

(1.40 高一点点)

(1.55 一定要买)

(1.56 好 那我一定要买)

etc.

Also, I randomly looked in a CCTV-dubbed drama and immediately found examples of ending 3rd tones without rises:

钻石情缘16

(At this stage, I'm thinking rises are used for emphasis...)

[doraemon]

It can be bit of a pain trying to pronounce two consecutive third tone characters. I've heard from my teacher that when this is the case, the first character becomes second tone (although I don't really do this since I haven't heard anyone else say that apart from her).

[creamyhorror]

It can be bit of a pain trying to pronounce two consecutive third tone characters. I've heard from my teacher that when this is the case, the first character becomes second tone (although I don't really do this since I haven't heard anyone else say that apart from her).

3-3 becoming 2-3 is a standard tone sandhi rule - just check any guide to Mandarin.

I think that clip certainly shows the difference in the stressed vs. unstressed tones (especially the third), as I mentioned in one of the earlier posts, going off this info about the crazy nature of third tones.

Nice link! That's a good rundown of the reality of tones and stress.

Edited April 9, 2010 at 11:06 AM by creamyhorror

[renzhe]

I think we all agree that the third tone is not always pronounced in the canonical way, especially in the middle of words, or when squeezed into very fast passages. After all, tones are just an approximation anyway, as wushijiao's link nicely shows.

What I find puzzling is the idea that the third tone NEVER sounds like it is taught, which goes contrary to my experience. In my experience, when it is alone, or at the end of a word or sentence, it tends to be full, and it is always full if you are putting stress on it (like the "好!" example).

It would be interesting if this really were the case in some regions.

[Glenn]

3-3 becoming 2-3 is a standard tone sandhi rule - just check any guide to Mandarin.

In my (limited) experience that isn't quite the case. It seems like shallow3-3 more than 2-3 at times. Other times it sounds like low tone-3. I think that 2 is just an approximation of what's probably already an approximation of a sound.

And I agree with renzhe -- I've definitely heard the rise at the end like it's described. But I mostly listen to 静雅思听 (and of course, as I was writing this, I just heard a 手 drop low and sit there at the end of a sentence).

[creamyhorror]

What I find puzzling is the idea that the third tone NEVER sounds like it is taught, which goes contrary to my experience. In my experience, when it is alone, or at the end of a word or sentence, it tends to be full, and it is always full if you are putting stress on it (like the "好!" example).

It would be interesting if this really were the case in some regions.

I don't find it very puzzling, since we hardly use it in Singapore. When learning pronunciation and throughout school my teachers never got us to add a rise. But we're definitely not the cultural center of Mandarin

I hadn't taken close notice of the mainland 3rd-tone rise until now. There's no doubt the rise exists and is sometimes/often deployed by mainland speakers on terminal 3rd-tones. I didn't think of it in terms of being the "original" tone; I just thought of it as an accent thing, in which speakers would sometimes sound the tone differently for emphasis or effect or simply just because it's their accent. (The change reminds me of the sandhi changes of phrase-terminal tones in Hokkien.)

Thinking back, the rise is probably something I'd unconsciously do if trying to mimic a Beijing accent. But I wouldn't use it in normal speech...sounds too 'fake-proper' for a speaker like me Maybe after I spend some time in China!

In any case, I think it's perfectly acceptable to use pure-low 3rd tones most or even all of the time, assuming one's goal is to be understood. I don't think it'll detract from how natural you sound, and it certainly won't hurt your intelligibility.

[Shi Tong]

I'm even more puzzled that there are people who have been learning for 9 years and have never heard a native speaker use it in speech.

Never is a bit over the top.

For example, one very notable exception is ni2hao3, which most people rise at the ends. That said, it applies to some and not others, and a lot of people still say a flat 3rd tone for this.

I have another clip of a soap opera from Taiwan, which shows you exactly how Taiwanese people speak Mandarin, which is very clear at the start. Note that the girl and the man speaking at the very beginning really dont raise their 3rd tone at all:

here

It's hard to make general statements on tones based on rapid-fire 200 wpm news snippets because tones come so fast that they tend to blend into each other. The newscaster Shi Tong posted has such a staccato accent that there's no time to pronounce the full 3rd tone even if she tried.

While I agree with this statement, I would also like to point out that most people speak this quickly in native tongues. This may be the reason for/ answer to this point.

Shi Tong's notion that the 3rd tone might stay consistently low in Taiwanese Mandarin is interesting.

Thanks.. I would say that it occurs most of the time when people are speaking naturally. If they're trying to stress the tone, then maybe the rising at the end would occur, and I would assume this would be at the ends of sentances with the 3rd tone final character.

What I find puzzling is the idea that the third tone NEVER sounds like it is taught, which goes contrary to my experience. In my experience, when it is alone, or at the end of a word or sentence, it tends to be full, and it is always full if you are putting stress on it (like the "好!" example).

Another thing to consider is Taiwanese people's obsession with end particles, which may also help to explain this phenomenon- You would barely ever hear anyone in Taiwan just say "hao3", and a lot of the time, when it ends in a 3rd tone, they would add an end particle, which would be a qingsheng- like "haoba, haola, haoe, howa" etc etc. This means that even when you have a 3rd tone end, you would probably find the qingsheng particle which affects the 3rd tone pronunciation.

I hope this spreads some further light.

I would also point out that this is not something which ALWAYS happens, simply something which occurs very frequently, especially in fast spoken language.

As creamyhorror just said, a rise in the 3rd tone would sound very proper, and would be like someone doing a beijing accent impression in Taiwan. Though it does occur amoung lots of people and especially in some groups, I would say most Taiwanese people speaking normally wouldn't raise the tone.

[Hofmann]

Uhh...

Part of the standard is that the third tone is low except at the end of an utterance. See? This is stuff you should have learned or been taught within the first few hours.

[Shi Tong]

Strangely NOT.. no.. which is very odd.

I basically found out that the rise at the end of the 3rd tone (which is supposed to appear always, according to my teaching (bair in mind 3 months)) only appears at the ends of phrases or sentences by osmosis.

That said, I still think that 3rd tone is so constantly spoken with a low flat tone, even when saying things like Hao3 on their own, that it's pretty ubiquitous.

hmm.. interesting.

[Altair]

Are we really talking about a "low flat tone" or a low falling tone? I think the tone should always drop a little bit and not actually stay level.

[Shi Tong]

AFAIC, a DROP in tone indicates a 4th tone ONLY, whereas a slight raise in tone at the end of a sentence is a 3rd tone. A second tone has a rising tone.

3rd tone, if in the middle of anything, should always be a level low tone without inflection.

[Altair]

According to what I understand, a 4th tone is a fall from high to low, represented as a drop from "5" to "1." Also, the volume goes goes abruptly from loud to soft, like a barked command in English. The half 3rd tone goes from medium low to low, represented as a drop from "2" to "1." Unlike with the 4th tone, the volume tends to increase as it drops toward a "creaky" voice. This Wikipedia page says:

# Third tone (low or dipping tone, 上聲/上声 shǎngshēng or shàngshēng, literal meaning: "up tone"):

has a mid-low to low descent; if at the end of a sentence or before a pause, it is then followed by a rising pitch. Between other tones it may simply be low.

I would agree with this, including the last statement about the 3rd tone sometimes being simply low, except that I think "rising tone" is a better translation of 上声 than "up tone" and that the final rise in pitch is actually somewhat optional at the end of a sentence.

I do not know much about Taiwanese or Min dialects, but I think Cantonese has a tone that is very similar to the half Mandarin 3rd tone which is called 陽平 (fourth tone?). It is used in words like 麻 and 時/时. I think this tone has free variation between a very low level tone and a low falling tone. Maybe Taiwanese has something similar that might affect some peoples pronunciation of the Mandarin 3rd tone.

[Shi Tong]

has a mid-low to low descent; if at the end of a sentence or before a pause, it is then followed by a rising pitch. Between other tones it may simply be low.

Well, exactly- it's a tone which stays low and flat between words.

I've never ever noticed a "dip" in the approach to the start of the third tone in my life, though I would certainly say that at the ends of sentences it can (sometimes) go up a bit. I suppose a percieved dip at the start of a third tone might appear if one is already speaking another tone which is either rising (2nd) or high (1st).

Apart from that, it is in my experience, very rare to hear a dip like the traditional pictograph of "how the third tone is supposed to sound".

[Altair]

I've never ever noticed a "dip" in the approach to the start of the third tone in my life

Listen to 你去过北京吗 on this page of nciku.com. At least to my ear, there is a clear drop in pitch over 北. This is despite the fact this third-tone syllable appears between other tones.

[roddy]

Third tone starts low, goes lower, stays there for a while, and maybe now and then comes back up. I'll wait for any alternative views to filter through into textbooks before I pay too much attention.