?

Log in

No account? Create an account

Previous Entry | Next Entry

Intonation in Chinese

weather: clear
outside: 21.6°C
mood: happy
No, I will NEVER run out of infinitesimally detailed things to say about the Chinese language. =) From linguaphiles:

Chinese people don't just use tones - they use intonation as well in much the same way we do...so in the class room, the phrase "Ming bai?" (are you clear?) can be answered with "Ming Bai." as a statement because you use different intonation patterens [sic].

I think I should try to explain a few things before I start...

A phoneme is the basic unit of human speech. It very roughly corresponds to a letter in the English alphabet, although there are cases where one letter is actually multiple phonemes (a long 'a') and a phoneme is represented by multiple letters ('ts', 'ch').

You can have voiced phonemes ('v'; vocal chords vibrate) and unvoiced phonemes ('f'; vocal chords do not vibrate, unless you're saying it wrong, like "fuh" or something).

The voiced phonemes have attributes associated with them: voice onset, voice segment and voice offset, among many others. The voice onset is the point in time where the vocal chords begin to vibrate (I have a little piece of useless trivia about Voice Onset Time, see below). The voice segment is the middle part, where you're holding the sound. It can be long, short, medium, just right, etc. And the voice offset is the trailing end.

Now.

We know that Chinese is a tonal language. We also know that there is a "second tone" which is a rising tone. To make this tone quality, you would make your voice rise, much like you would when you're asking a yes-no question in English.

How, then, do the Chinese differentiate between a second tone that is asking a question versus a second tone that is making a statement? That would be the 64 Million Dollar Question of the day... well, of this post, anyway =)

For example, the phrase 明白 (míng bái), ends with a second tone. As a question, it's "明白﹖" As an answer (which, in this case, is a statement), it's "明白。"

It's very very subtle, but there is an intonation portion at the voice offset of "bái".

In the question, the quality of the second tone is tending toward a continued rise as if your voice were going to follow through with the implied interrogative particle "嗎" (which is a high neutral tone that sounds very similar to the first tone). Example: 明白嗎﹖ (míng bái mā?).

In the answer, the second tone quality is as if your voice were going to follow through with the implied perfect/completive/modal particle "了" (which is a mid, steady neutral tone if it follows a second tone). Example: 明白了。 (míng bái le).

If you wanted to get technical about it, this alternate second tone is a separate tone. It is distinct and it has meaning assigned to it.

When people say that Mandarin is easier because it only has four tones and Cantonese has seven, that's not true. Cantonese only has seven because they've decided to differentiate seven of them. If we differentiated all the different tones in Mandarin, it's on par.

Mandarin has many many more tones than just the four. There are (linguistically speaking) at least 8: the four that you're taught, I count at least three different flavours of the neutral tone and this alternate second tone. I'm positive there are more of these subtle ones that come up in different contexts and circumstances that are lumped into one of the regular four. In fact, I know there are ones that waver in intonation, we purposely fluctuate the tone a little.

Native speakers will tell you that's not true and that there are only four tones. Some will deny that even one neutral tone exists. Many certified Mandarin teachers don't understand stuff like this. My mother has been teaching adult and children's Mandarin and Cantonese classes for over 40 years and she is still finding out in dribs and drabs that I was right about a lot of these things.


Little Piece of Useless Trivia

The consonant 'b' sound does not actually exist in English. Neither does the 'd', for the same reason.

Because the voice onset time is so late, what we're really saying is an unaspirated 'p' for 'b' and an unaspirated 't' for 'd'. The English 'p' and 't' are aspirated 'p' and 't' which means we put an extra puff of air into it. An English 'b' is a 'p' without the extra puff of air, likewise for the 'd'.

In many European languages (and you Europeans on my list, correct me if I'm wrong), a 'b' is actually a 'b'; the voice onset is much earlier than the English 'b'. At least, I think it's different in French. "quatre bières" versus "four beers" sounds different to me.

My highschool Physics teacher, Evan J., spoke Afrikaans which is very similar to Dutch. He would always say 'b' and 'd' on words that begin with 'p' and 't'. It was fun talking about "backets of energy", "photon dorbedos" and "Desla goils". =)


Comments

( 20 comments — Leave a comment )
yueni
May. 29th, 2005 05:31 am (UTC)
I find that when I'm saying "明白?" I always add the "吗" at the end. Curiously enough, I've just recently realised that if the second tone is uttered twice in a row like in 明白, the second one sounds like the third tone, if that makes any sense. I was always taught the four tones when in school, and it was only when I moved to the US that I even heard about the "fifth" tone.

I do think you are right. There are four "official" tones, but speakers tend to differentiate them minutely when speaking. There are all sorts of tonation rules that I do without even being aware of them. One of the ones that I was explicitly taught was that when two third tone words were together, the second one was pronounced with a rising tone (i.e. the second tone). I wish I were more awake to do a more lucid discussion about this. But this is really interesting.

By the way, how do you put in the tonations for the hanyu pinyin on your vowels? I've been trying to figure it out, but I can never do it right.
bride
May. 29th, 2005 05:41 am (UTC)
the second one sounds like the third tone

It's a mid neutral tone, and I think it only appears if you say two characters in isolation or just before a pause (a comma in the sentence). If you had them in a sentence, it would be two full second tones. If there were three in a row, the third one would be neutralized.

One of the ones that I was explicitly taught was that when two third tone words were together, the second one was pronounced with a rising tone (i.e. the second tone).

The first one becomes the second tone =) 你好, nǐ hǎo => ní hǎo. =)

how do you put in the tonations for the hanyu pinyin on your vowels?

I copy and paste them from NJStar.
yueni
May. 29th, 2005 05:08 pm (UTC)
The first one becomes the second tone =) 你好, nǐ hǎo => ní hǎo. =)

*doh* Yes. That. I meant that. *hangs head* Note to self: Do not reply when tired and incoherent.
kalev
May. 29th, 2005 06:36 am (UTC)
I know French and I'm not sure it's different. To me, when I say "bonjour" (French, évidement), it's the same as when I say "bye" (English, naturally).

And when I do the alphabet in both languages, I'm pretty sure it's the same lip movements/breathing. I mean, French bee is pronounced somewhere between bey and bay but the actual consonant seems the same to me.

What is "voice onset time," anyway?
bride
May. 29th, 2005 07:16 am (UTC)
What is Voice Onset Time? ... I can't find one that's less geeky than that. =P
bride
May. 29th, 2005 07:21 am (UTC)
Oh, I found this one. If I'm reading that correctly, it looks like French has an earlier VOT than English, which is why the voiced consonants sound different to me (because they are, ha!) =)
bride
May. 29th, 2005 07:25 am (UTC)
Okay, one more and I'm going to shut up. Take a look at the chart and caption at the bottom.
kalev
May. 29th, 2005 05:10 pm (UTC)
This one is the best. But this kind of micro-thing drives me mad, actually, which is weird because normally I like knowing ticky-boo stuff. But when I start thinking about pronunciation on such a small scale, I get all fucked up and cannot speak. I'm just happy I learned French when I was 5 and so can just speak it without trying to think about how to speak it, and that knowing it will hopefully make it easier to pick up any new languages I might one day want to speak or know.

Some things are better understood instictively than consciously.
bride
May. 30th, 2005 01:34 am (UTC)
I'm just happy I learned French when I was 5 and so can just speak it without trying to think about how to speak it

And that truly _is_ the best way to go about learning a language - to grow up with it, speaking it all your life =D

But it's also amazing to me to pick it apart and realize why people have the FOREIGN ACCENTS that they do.
science_vixen
May. 29th, 2005 12:17 pm (UTC)
Actually it is.
The 'b' in the english bye is a bit slow, where you roll your lips outward while creating the 'b' sound. The French 'b' in bonjour is a bit more explosive, unless you try to be sultry that is. it's more like a 'p' without touting your lips.
The same goes for Dutch and German.
pne
May. 29th, 2005 03:01 pm (UTC)
phones vs phonemes
If you're going to get geeky, I think that when you were talking about "phonemes", you really meant "phones".

For example, with my (limited) understanding of linguistics/phonology/whatever, A phoneme is the basic unit of human speech. doesn't make sense -- phonemes only exist within the context of a given language. Phones, on the other hand, are unrelated to language and are objectively-measurable portions of human speech in general.

there are cases where one letter is actually multiple phonemes (a long 'a')

This bit also seems odd, since to me, "long 'a'" is clearly one phoneme in English -- "hat" and "hate" are both exactly three phonemes long, though the second could be analysed as four phones (I'm not sure how diphthongs/glides count in terms of number of phones).

a phoneme is represented by multiple letters ('ts', 'ch').

Here, on the other hand, you appear to be counting affricates as one phone.

In the context of English, I don't think there's a phoneme 'ts': 'hats' is four phonemes for me, regardless of the phonological representation. 'Ch', on the other hand, is a phoneme in English. (Not sure about German, for example; it's probably two phonemes there, 't' + 'sh', and occurs mostly in loanwords and onomatopoeia.)

You can have voiced phonemes ('v'; vocal chords vibrate) and unvoiced phonemes ('f'; vocal chords do not vibrate, unless you're saying it wrong, like "fuh" or something).

This bit, to me, is definitely about phones -- since there are languages where phonemes can have both voiced and voiceless allophones. They may be positional (depending on the surrounding context) or even in free variation (i.e. pronouncing a hypothetical basic word */pat/ as [pad] or [bat] would mean exactly the same thing to a native speaker of the language). In such languages, phonemes may not have explicit voicedness. (For example, in Niuean, I believe the only explicit voiced/voiceless contrast is between [f] and [v], and [v] probably comes from original *[w], giving no original/native specifically-voiced phonemes.)
bride
May. 29th, 2005 04:07 pm (UTC)
Re: phones vs phonemes
I think that when you were talking about "phonemes", you really meant "phones".

*hangs head* Yeah, I keep doing that. =P
timwi
May. 31st, 2005 06:09 pm (UTC)
Re: phones vs phonemes
I would say that "tsch" is a phoneme in German, not two. Consider "Quatsch" and "Tschüß". Even "Tschuldigung", while clearly deriving from "ent"+"schuld" (etc.), I think is generally perceived to begin with "tsch" and not with "t" followed by "sch".

Other than that, I fully agree with everything you said. But there are a few more things I would like to add. :-)

bride said that Mandarin actually has more than four tones even though only four are taught in normal language classes. When talking about "tones" as an analogue to "phones", this is probably true; but then English has loads of tones too, even though nobody classifies Engilsh as a tonal language. The secret here is the concept of a toneme, the analogue to the phoneme: Mandarin has four tonemes, English only one. Cantonese has seven, and thus more than Mandarin.

Furthermore, bride mentioned that some Mandarin speakers will even deny the existance of the fourth toneme and claim there are only three. Well, this nicely mirrors the fact that most Germans believe there are only seven "vowels" in German, a e i o ö u ü, when in fact there are at least 15 vocalic phonemes, [aː a eː ɛ ǝ iː ɪ oː ɔ øː œ yː ʏ]. I know n_true would add [ɐ] as a 16th.

Now, of course you can say that German orthography influences people, because when they say "there are seven vowels", they probably mean "we use seven different symbols to represent our vowels in writing". But orthography doesn't determine phonemes. They might think of ü as "one vowel", but they will still clearly distinguish between [hyːtǝ] and [hʏtǝ], even though the only pronounced difference is the vowel, and it's written as ü in both words (Hüte and Hütte).

What I'm saying essentially amounts to stating that the concept of phonemes is far from intuitive. It cannot be determined from what average native speakers say, but it also cannot be determined from purely objective measurements (because then you'll just be measuring phones, not phonemes).
timwi
May. 31st, 2005 06:24 pm (UTC)
Re: phones vs phonemes
Of course, in the list of German phonemes, I forgot [uː ʊ]. :-)
pne
Jun. 1st, 2005 05:59 am (UTC)
Re: phones vs phonemes, tones vs tonemes, German
The secret here is the concept of a toneme, the analogue to the phoneme

*nods* This also takes care, for example, of those Chinese dialects where syllables can have different tone contours in isolation or around certain other syllables (tone sandhi) -- Mandarin has this for two third tones next to one another, but I've heard that there are dialects with more complex sandhi rules, and I'd say those are simply "allotones" of the same toneme rather than separate tones.

there are at least 15 vocalic phonemes, [aː a eː ɛ ǝ iː ɪ oː ɔ øː œ yː ʏ]

(Shouldn't those be /slashes/ around the list rather than [brackets], since you're talking about phonemes?)

Interesting that you didn't count /ɛː/, which is a marginal phoneme in my idiolect; standard German /ɛː/ usually merges with /eː/ for me into [eː], so "Reeder" and "Räder" sound the same. But I suppose if I speak very carefully or want to emphasis the difference, I might say [ɛː] but only in words where standard German has the phoneme /ɛː/, so it might be a separate phoneme. In general, though, I don't make a phonemic distinction.

I know n_true would add [ɐ] as a 16th.

As distinct from /r/ or /ɛr/? Do you know whether he has a minimal pair for /ɐ/ vs. "the usual cases" where standard German has [ɐ]?

But orthography doesn't determine phonemes.

*nods* And sometimes it doesn't even mark separate phonemes, e.g. "weg" vs "Weg", which are something like /vɛg/ vs /ve:g/ for me (and, phonetically, [vɛk] ~ [vɛç] vs [ve:k] ~ [ve:ç]), or "Lache" (way of laughing) vs "Lache" (puddle).

What I'm saying essentially amounts to stating that the concept of phonemes is far from intuitive.

Definitely agree. I thought I had a decent concept on the idea of phonemes, but I thought for quite a while that [ʌ] and [ə] were separate phonemes in English, while they're probably merely allophones of one phoneme /ə/ depending on whether they're in a stressed or an unstressed syllable; however, they "felt" like separate sounds to me. Similarly with German [x] and [ç], which "feel" rather different to me but may be allophones of a single phoneme. (I'm told that whether they're one phoneme or two depends on which Germanicist you ask, and that there's no consensus yet, though minimal pairs tend, AIUI, to involve the suffix -chen which always has [ç] and usually, but not always, umlauts the preceding syllable; the classic near-minimal pair for me is "Frauchen" vs "fauchen", though I've heard *"Kuhchen" vs "Kuchen" as well, but *"Kuhchen" is doubtful for me.)

On the other hand, [θ] and [ð] are definitely separate phonemes in English, at least in my idiolect, even though some people consider them allophones -- perhaps because they have the same orthographic representation, <th>. But I have minimal pairs e.g. in "thigh" vs "thy".
timwi
Jun. 1st, 2005 02:09 pm (UTC)
Re: phones vs phonemes, tones vs tonemes, German
We've strayed a bit from the original topic, but anyway: Yes, I forgot about [ɛː], which is separate in my dialect from [eː]. Nevertheless, I am aware that there are dialects that merge them. I wasn't sure if Standard German does, but I know that the German-dubbers of Walt Disney's Gummi Bears do, hence Gummibären (Gummi bears) and Gummibeeren (Gummi berries) are indistinguishable! Quite frustrating. :)

Also thanks for explaining the difference between square brackets and slashes: I never knew that and always wondered! :-)
bride
Jun. 1st, 2005 03:51 pm (UTC)
Re: phones vs phonemes, tones vs tonemes, German
Heh, I'm having a blast following along. Feel free to stray any which way on this stuff in my journal =D
pne
Jun. 1st, 2005 05:10 pm (UTC)
Re: phones vs phonemes
Also thanks for explaining the difference between square brackets and slashes: I never knew that and always wondered! :-)

That difference (/slashes/ for phonemic representation and [brackets] for phonetic representation) is what I learned; it seems to be fairly standard.

Strictly speaking, brackets can be used for broad or narrow phonetic representation, though; for example, one might write [pɛn] for English "pen" and ignore the aspiration, or write [pʰɛn] and include it.

I'm not sure whether there's a standard for specifying orthographic representation, but I've seen <angle brackets> used that way -- e.g. "The word <lead> can be pronounced either /lid/ or /lɛd/".

Finally, I've usually seen IPA used for transcribing phones, but for transcribing phonemes (which are inherently language-specific), it seems that some people prefer other systems, though I've also seen IPA used for that. (For example, what I'd write as /it/ and /ɪt/, for <eat> and <it>, I've also seen written /iyt/ vs /it/.) And length is sometimes left out of phonemic transcriptions if length is not, indeed, phonemic (as with /it/ vs /ɪt/, which I'd write more phonetically as [i:t] vs [ɪt]).
bride
Jun. 1st, 2005 03:52 pm (UTC)
Re: phones vs phonemes, tones vs tonemes, German
timwi says:
We've strayed a bit from the original topic, but anyway...

I'm having a blast following along. Feel free to stray any which way on this stuff in my journal =D
timwi
May. 31st, 2005 06:23 pm (UTC)
In addition to what I already said above, I would like to comment on your comparison between the English and French b as well. :-)

I don't think they are pronounced differently as such, but there is something else besides voice onset that makes them sound different nonetheless. English employs glottal stops, French doesn't. When you say "bye", you are probably pronouncing a glottal stop at the same time as the b itself. A French would instead pronounce the b with their glottis opened.

The difference is particularly noticeable in words that begin with a vowel. In French they are pronounced in such a way that, to English speakers, it often sounds like there's a very soft h-like sound before the vowel. In fact, when speaking English, Frenchmen are often unable to pronounce any difference between pairs of words such as "all" and "hall", or "ear" and "hear".

When you say "four beers", you are also pronouncing a glottal stop. Even though the r is vocalic and the b is voiced, you will close your glottis in between, stopping the airflow and thus the voice for a short moment. When saying quatre bières, a Frenchman will not stop the airflow except when closing their lips (which doesn't stop the voice).

You are right that in English (and, in fact, in German), t and p are aspirated while their voiced counterparts are not. However, it is not true (as you assert) that this is the only difference between them. After all, the French do not aspirate d or t, and as a result a German or English speaker would probably easily mishear "tant" as "dans", but no French would ever assert that "dans" and "tant" are pronounced the same.
( 20 comments — Leave a comment )

Profile

eLouai
bride
The Bride of the First House

Latest Month

March 2015
S M T W T F S
1234567
891011121314
15161718192021
22232425262728
293031