Lip shape for Zh, Ch, Sh

[Hofmann]

The vowel in 之 and 資 are different. Try to get someone to say 這 and 仄.

[陳德聰]

I feel like I'm channeling skylee right now but Cheng Yangyang being a native Cantonese speaker doesn't change the fact that she has impeccable Mandarin. I browsed through her other videos and she doesn't even have any semblance of what I like to call "old HK businessman" accent.

 

As for 之 vs 资, I'd always assumed they were both barred i, are you saying that 资 is barred i and 之 is barred u?

 

Working on finding a clip with "ze".

[Hofmann]

One can transcribe both phonemically as /ɨ/, but in 資 one's tongue is in the alveolar position and 之 in retroflex. A narrow transcription would be [z̩] and [ʐ̩].

[陳德聰]

But then if you contend that the rounding comes from the round or unround quality of the vowel, isn't it a bit ad hoc to say that the reason there is rounding in retroflex is that the vocalic consonant is just round? How does that explain roundness in this context and lack thereof in others?

[Hofmann]

No. Assimilation exists in many languages, including Mandarin, and not only in this instance.

[陳德聰]

I'm not contesting that vowel roundness is ripe for assimilation, I'm saying that if you are proposing a narrow transcription of 资 as vocalic [z] (btw how is that an accurate narrow transcription?) and 之 as vocalic [ʐ], then your analysis requires that your underlying form for the retroflex series be rounded... Which would be fine, but then you'd essentially be saying that the retroflexes are round, and lose their roundness when followed by unround vowels. I would be okay with that, but then I would want to know why the retroflexes are underlyingly round when all the other sibilants aren't.

[Hofmann]

Whether or not the retroflex series is rounded is irrelevant because they never occur isolation in Mandarin. They are always followed by some vowel, and most of them are not rounded. It seems the OP wanted to ask whether or not retroflex series when followed by [ʐ̩] is actually labialized, perhaps because the vowel is also labialized [ʐ̩ʷ]. (This transcription looks like it's a syllabic fricative. There's no better way to reflect that it's an approximant.) I'd say no until someone shows me it is.

 

As for the other vowels, as I said before, it depends on the quality of the vowel.

[陳德聰]

I think maybe the reason I misunderstood you is that you initially proposed that the only relevant issue is vowel round quality, but then in #23 you don't give [ʐ̩ʷ], you give [ʐ̩]. I also thought you were transcribing the entire syllable rather than just the vowel.

 

From #27, it's much more clear what you're trying to say. But then I am totally thrown off by

whether or not retroflex series when followed by [ʐ̩] is actually labialized, perhaps because the vowel is also labialized [ʐ̩ʷ] ... I'd say no until someone shows me it is.

Doesn't that directly contradict what you just said about assimilation? I'm not trying to push retroflexes being labialized in every context, I'm just trying to wrap my head around what seems like some conflicting information that you're providing.

 

If the vowel is [ʐ̩ʷ], then according to you the proceeding retroflex would be labialized due to assimilation, wouldn't it? If what you're saying is that you don't believe the vowel should be transcribed as [ʐ̩ʷ] because you haven't seen it labialized, then I don't really know what to say.

[Hofmann]

I don't think it is labialized.

[陳德聰]

K so... what does your analysis say about the clips that were posted in #17 and #18 then? Neither the consonant, nor the vowel in those cases are labialized to you? I can't tell if you're trolling or what.

[Hofmann]

Not labialized. Labialization looks like /u/ or /y/.

[陳德聰]

So you're trolling then.

[Hofmann]

Alright, I'm back. Totally not trolling. I don't know what the problem is. The first answer to the OP was correct even if it wasn't very credible looking. Labialization isn't a feature of the retroflex series. The slight rounding you see in the videos isn't necessary, nor is the rounding referenced in the paper you linked to. When I say labialization looks like /u/ or /y/, did you even look at those? Heck, you can even try (if you speak General American) saying "red," as /ɹ/ is labialized [ɹʷ].

[陳德聰]

When you said labialization looks like /u/ or /y/, an inadequate definition of labialization, I thought you were being willfully ridiculous because you usually display impressive linguistic knowledge.

 

Edit: As far as I knew we had already moved beyond whether or not it was optional.

[TechKatz]

Hello everybody,

 

First post here. English is not my first language so please bear with me.

When I started to learn Mandarin I also couldn’t wrap my head about the s/z/c and sh/zh/ch initials so I did some research to understand them better.

 

First of all s/z/c and sh/zh/ch are all Sibilants. These are made by directing a stream of air with the tongue towards the sharp edge of the teeth. The teeth are held close together. If the lips were relaxed, the mouth would be closed.

 

 

Pronunciation in the Pinyin alphabet:

 

The Initials in the Pinyin alphabet are always pronounced a certain way. The consonants are never pronounced in isolation but are neighbored by a certain vowel. It’s the same with the English alphabet: for example b is always pronounced [bi], f is always pronounced [əf] and so on. The s/z/c and sh/zh/ch initials in the Pinyin alphabet are actually pronounced as si/zi/ci and shi/zhi/chi (with high tone).

 

The sound represented by the letter i in these two series is somewhat tricky to describe phonetically. This is because it has both consonant and vowel features. It is also different for the s/z/c and the sh/zh/ch series. What is basically happening is that the tongue stays close to the articulation point of these consonants and the sound is only slightly altered.

 

A possible description for the sound following the alveolar/dental series s/z/c [s/ts/tsʰ] is z̞̍^ɨ (So si/zi/ci in Pinyin would be [sz̞̍^ɨ /tsz̞̍^ɨ /tsʰz̞̍^ɨ] in IPA).

Because this sound forms the nucleus of the syllable and is an only slightly altered consonant it can be regarded as a syllabic consonant (can be marked as vertical line above or below the letter: s -> s̍). While [s/ts/tsʰ] are all voiceless the following syllabic consonant is voiced (s̍ -> z̍). Furthermore the tip of the tongue will usually be slightly lowered away from the articulation point so there is no more friction (z̍ -> z̞̍). The rest of the articulation features of the initial Sibilant alve​olar fricative are kept (nearly clenched teeth, groove in the back of the tongue and so on). The superscript ɨ (unrounded, close, central) is added to roughly describe the corresponding vowel based on the relative position of the tongue to the roof and back of the mouth. This is a bit tricky because when pronouncing vowels the tongue has usually no groove and the tip of the tongue is not so close to an articulation point. It’s noticeably different from the i-sound in xi/zi/qi [ɕi/tɕi/tɕʰi] because here the tongue is flat throughout (palatal). Furthermore the ɨ is there to show the unrounded vowel quality. For high vowels the unrounded version is pronounced with spread lips (see the pictures for s/z/c in post #17).

 

The vowel/syllabic consonant following the consonants in the retroflex series sh/zh/ch [ʂ/ʈʂ/ʈʂʰ] can be described as ʐ̍˕^ɘ (So shi/zhi/chi in Pinyin would be [ʂʐ̍˕^ɘ/ʈʂʐ̍˕^ɘ/ʈʂʰʐ̍˕^ɘ] in IPA).

The steps are same as above: ʂ -> ʂ̍ (syllabic consonant) -> ʐ̍ (voiced) -> ʐ̍˕ (no friction). The result is actually the sound for the Pinyin letter r which is part of the retroflex series as well (can also be pronounced with friction: ʐ̍). The corresponding vowel would be something close to ɘ (unrounded, close-mid, central). The difference to ɨ is the lower height of the tongue. When switching from si to shi for example it can be felt that while the tip of the tongue curls back the rest of the tongue drops quite a bit. This vowel is also unrounded. And here lies the explanation for the lip shape seen in post #17 (sh/zh/ch). In order to create an opening of the same size as for ɘ without lowering the jaw (teeth need to stay close together to keep the sibilant feature) the upper lip is pursed while the lower lip is protruded. Compared to z̞̍^ɨ the lips are more apart and less stretched (can also be seen in post #17 or the video in post #18). This corresponds to the difference between the normal ɨ and ɘ vowels.

 

 

Lip shape before unrounded vowels:

 

The lip shape before unrounded vowels matches the above described shape for si/zi/ci and shi/zhi/chi. However, it should be noted that the lip shapes seen in post #17 are rather extreme because they require quite some effort. In normal speech the lips are more relaxed to the point that the above described lip shapes are barely noticeable. This can also be observed in English words that start with a sibilant consonant and are followed by an unrounded vowel, for example “see”, “zip”, “jeep”, “sharp”, “chirp” – compare the lip shape when pronouncing them deliberately with the lip shape when pronouncing them casually.

 

 

Lip shape before rounded vowels:

 

Before rounded vowels s/z/c and sh/zh/ch are labialized. The lip shape for s/z/c corresponds with ʉ vowel (the rounded equivalent of ɨ) and the shape for sh/zh/ch roughly corresponds with the ɵ vowel (the rounded equivalent of ɘ). Rounding is more natural for high vowels due to the lips being closer together. That’s why the labialization of the syllabic consonant is present not only in deliberate speech but in normal speech as well.

 

Hope this was more clarifying than confusing. Some of the things I described are based on my own observation so feel free to correct me.

[Guest Mira Price]

Here's a neat video pinyin chart I found today that was helpful to me.

 

It has video tutorials for all those hard to pronounce Mandarin sounds like zh, ch, sh: you'll be able to see a native Chinese speaker's lip shape very clearly: <Admin Note: Gratuitous link to own content removed>

[imron]

Here's a neat video pinyin chart I found today

You just found it?  While browsing your company's website?  Wow, what a coincidence