Talk:Standard Chinese phonology

Tilde in Phonetic Transcriptions
"", "", "[ ~ ]", "[ ~ ]", "", "",... What is this tilde supposed to mean? There is no explanation anywhere on this site nor on the wikipedia website on phonetic transcription and I could not find anything on any other website using a search engine. 138.246.2.9 (talk) 18:13, 24 June 2018 (UTC)


 * It's a standard way of indicating allophonic variation. For example, the phoneme /x/ can be pronounced either [h] or [x]. Zgialor (talk) 20:21, 22 September 2018 (UTC)


 * Many different uses of ~ in many fields of study are explained in the Tilde article, but this one is missing. I wonder which of our articles explains ~ as a symbol indicating linguistic variation. Love —LiliCharlie (talk) 22:21, 22 September 2018 (UTC)
 * I don't know if you did it or someone else, but the usage has been added under the header Linguistics on the Tilde page. Barefoot Banana (talk) 16:45, 10 September 2023 (UTC)

P vs B and /p/ vs /b/ vs /p^h/
For some reason, Wikipedia's phoneme list has /p/ as the pinyin b and /p^h/ as the pinyin p. I learned Mandarin with p as your regular English p (i.e. /p/) and b as very close to our regular English b (i.e. /b/). I get it, the Mandarin b is unaspirated and I can memorize that.

My question is: why is the Wikipedia phoneme list listing the Mandarin p as /p^h/ and the Mandarin b as a /p/ when it should be listing the Mandarin p as /p/? And maybe the Mandarin b as /p^h/? — Preceding unsigned comment added by Cleverwater (talk • contribs) 16:40, 3 May 2020 (UTC)
 * Take a look at English phonology to see an explanation of the sounds of English. In short, English p is and English b is often so devoiced that it might as well be . — Æµ§œš¹  [lɛts b̥iː pʰəˈlaɪˀt]  17:38, 3 May 2020 (UTC)
 * As I understand it, the Standard Mandarin "/b/" isn't really voiced except in some weak positions. This is what I've read anyway. So although it might be easier, when teaching English speakers, to tell them that it's a "/b/" (that will ensure that it's unaspirated, and any resulting voicing won't matter that much), it's not really an accurate description - it would be less helpful when teaching speakers of other languages whose /p/ might not naturally be aspirated, for example. W. P. Uzer (talk) 20:27, 3 May 2020 (UTC)

I am a native speaker and I think p in Chinese is same sound as/p/ and b same sound as /b/? I never seen anyone pronouncing b as /p/ in my life. Am I getting something wrong? — Preceding unsigned comment added by 89.19.89.103 (talk) 12:02, 31 December 2023 (UTC)
 * In English, at least US English, initial /b/ and /p/ differ in at least ways: /b/ comes out as a voiced unaspirated stop and /p/ comes out as a unvoiced aspirated stop. If what you are conscious of about the difference between English /b/ and /p/ isn't the voicing, then you might not notice that Mandarin /b/ and /p/ are both unvoiced. So Mandarin /b/ isn't realized as the same sound as English /b/. Largoplazo (talk) 12:39, 31 December 2023 (UTC)

Difficulty in using this page
I am studying Mandarin and finding this page too difficult to use.

The recordings of sounds are too difficult for me to use, because clicking them does not play the sound, but instead changes the page to a sound player. It is impossible to hear the sound while seeing the character and the Pinyin spelling, in relation to other sounds.

I can't follow all the technical words like "palatal" or the IPA transcriptions because, like most English speakers, I have no training in the classification of formal speech sounds or in phonetics.

Also, the Chinese characters for each sound are too small for me to see, and do not include all the characters that have the same consonant sound.

I need clear recordings, and I need them repeated along with both the Chinese and Pinyin characters clearly visible in large print, or at least a reference to this kind of tool if it already exists on the Web.

The characters I have the most trouble with as an English speaker are these:

1) xi and shi 2) ji and zhi 3) qi and chi, and I also have trouble with most dipthongs.

To those editors who would object that the information is already there and WP is not a language learning tool, I respond simply: Wikipedia is an encyclopedia; there is no exclusion for teaching any subject or for being a tool for learning. David Spector (talk) 16:24, 10 July 2021 (UTC)


 * ,, are written in pinyin with an  vowel, but in those syllables it's more of a syllabic fricative (alternatively analyzed as a central /ɨ/ vowel), [ʂʐ̩] [tʂʐ̩] [tʂʰʐ̩] You're basically shushing someone.
 * , ,  are most akin to English (the pronoun), (as in the letter G), (as in cheese). The articulation isn't the same, but it's probably easier for you to distinguish the retroflex and alveolo-palatals by vowel. Wandering Maiden (talk) 16:34, 11 January 2023 (UTC)
 * Wikipedia is indeed a tool for learning, and if you want to learn about phonology, which is a branch of linguistics, you will encounter linguistic terminology. From the five pillars page: Wikipedia combines many features of general and specialized encyclopedias. Lay people do not generally know the word "phonology", so I do not see any issue whatsoever with more specialised topics and presupposing basic linguistic knowledge. After an introductory course in phonology, most (and perhaps all) of the terminology here should be clear.
 * If you want information about pronunciation, it makes much more sense to search for videos online by native speakers than to go look on wikipedia. For Standard Chinese, there is an extremely large amount of learning material available specifically intended for language learners. Wikipedia shouldn't even try to be as useful for language learners as those websites already are. Barefoot Banana (talk) 16:39, 10 September 2023 (UTC)

Re: Vowels – Standard Chinese can be analyzed as having five or two vowel phonemes.[citation needed]
A good reference is https://www.researchgate.net/publication/291246853_Mandarin_Vowels_Revisited_Evidence_from_Electromagnetic_Articulography especially subsection 2.2. The authors cite sources for interpretations of two to six basic vowels for Mandarin. I would edit the page myself, but I'm not sure how to create a citation with this kind of source, e.g. whether to use the Journal or the URI.

20:45, 17 December 2021 (UTC) — Preceding unsigned comment added by 24.141.211.239 (talk)
 * Just added it. Erinius (talk) 16:03, 20 December 2021 (UTC)


 * Even the linked reference says 2-vowel interpretation is an extreme approach. It is not standard to treat /i/ /u/ /y/ as glides with no vowel nucleus or call any vowel with complementary distribution an allophone (when there are common minimal pairs with other vowels). Furthermore the authors very oddly transcribe 這個字 as /ʈʂɤ kə tsi/ when the first two vowels are always the same in pinyin and zhuyin, and the last vowel should be /ɨ/ or /ɹ̩/ or /z̩/ in accordance with the "Syllabic consonants" section. Their vowel chart is also strange since i have never seen /ɯ/ treated as a separate vowel.
 * So i do not think this is a good source. 202.94.70.52 (talk) 07:57, 20 July 2022 (UTC)
 * I think that this source is fine if it is used as a reference for the 2-6 vowel claim. The article in general, however, is indeed not of the highest quality and suffers from poor language editing.
 * /ʈʂɤ kə/ can be taken to have different vowels because  has a falling tone and  is toneless and reduced. The /i/ in /tsi/ is a phonemic transcription. The [ɯ] is used for the realisation of  in etc., and it is not strange to list it as a separate "vowel". But boy I have issues with that transcription. Barefoot Banana (talk) 16:26, 10 September 2023 (UTC)

Alveolo-palatal series and a supposed inherent [ʲ]
The section in question: "That is, syllables represented in pinyin as beginning ⟨ji-⟩, ⟨qi-⟩, ⟨xi-⟩, ⟨ju-⟩, ⟨qu-⟩, ⟨xu-⟩ (followed by a vowel) are taken to begin [t͡ɕj], [t͡ɕʰj], [ɕj], [t͡ɕɥ], [t͡ɕʰɥ], [ɕɥ]. The actual pronunciations are more like [t͡ɕ], [t͡ɕʰ], [ɕ], [t͡ɕʷ], [t͡ɕʰʷ], [ɕʷ]".

In other words, this statement says that the alveolo-palatals already have an inherent [ʲ] articulation that causes palatalization of the vowel, and so the palatal glide slot is not actually present. This feels incorrect to me and I disagree with this statement.

If we compare with Korean 차,자 /tɕ(ʰ)a/ and Japanese ちゃ,チャ /tɕa/, we can see that neither has an inherent palatalization of the vowel and Standard Chinese /t͡ɕjʰa/ does not sound the same as the aforementioned Korean and Japanese syllables (with regards to the palatalization of the vowel). In Cantonese, the /s/ is often very retracted (barring speakers that have a rounding-conditioned palatalization), giving its affricates /ts tsʰ/ a quality that is more akin to [tɕ(ʰ)a] of Japanese and Korean. I can physically articulate both the presence and absence of a palatalization of the vowel ʲV with this consonant quality, so in my opinion this statement needs to be amended. Wandering Maiden (talk) 16:24, 11 January 2023 (UTC)


 * There are indeed problems with this paragraph. "In phonological analysis, it is often assumed that" and then there is no source! I may try to find at least two later.
 * But I do not draw the same conclusion from the line that you quote. I would in fact interpret it as a medial slot being there obligatorily. And I would not necessarily disagree with that, based on what I know. I do agree that there is a difference between Japanese and Mandarin, but I don't think that transcriptions can be used as arguments. That's circular, don't you think? Barefoot Banana (talk) 16:12, 10 September 2023 (UTC)
 * @Barefoot Banana I'm not sure what you mean by the transcriptions being a circular argument. I bring up Japanese and Korean because despite bring transcribed the same, they are pronounced differently as far regarding vowel palatalization (pre-vowel palatal medial).
 * This indicates that either alveolo-palatals do not have an obligatory glide or that Japanese/Korean don't actually have alveolo-palatals. If the former, then there actually is a distinct medial [j] and the statement is incorrect.
 * What you're saying about an obligatory glide could be the case about phonologically representing it, so /tɕ(ʰ)-/ will always surface with a glide phonetically [tɕ(ʰ)j-]. But the line it's talking about the surface phonetic realization (it's using brackets []). The line states that phonetically the glide is already part of articulation of the alveolo-palatal consonant, which is what I'm saying is incorrect. Wandering Maiden (talk) 17:43, 10 September 2023 (UTC)
 * You transcribe Korean and Japanese as /tɕ(ʰ)a/ and Mandarin as /tɕ(ʰ)ja/. And you say that they are pronounced differently. But didn't you choose these different transcriptions because they sound different?
 * But if you meant that the article uses the transcription /tɕ(ʰ)a/ for Mandarin, same as for Japanese and Korean, then it's not circular to expect a difference.
 * Anyhow: "In phonological analysis, it is often assumed that, when not followed by one of the high front vowels [i] or [y], the alveolo-palatals consist of a consonant followed by a palatal glide ([j] or [ɥ]).". Whoever wrote this used square brackets, but actually it should be "/i/ or /y/" because it's about phonology and phonotactics, not actual realisations. I think this makes it ambiguous whether they are talking about phonetics or phonology with [j] or [ɥ].
 * Because this paragraph is unsourced anyway, I'll look up what Duanmu and Lin write about this and make changes accordingly. Barefoot Banana (talk) 19:22, 10 September 2023 (UTC)
 * Duanmu (2007) writes that palatals are consonant+glide combinations. He only uses square brackets so it's very unclear, but he means that e.g. /si/ is realised as [sji] or [ɕi] depending on the speaker.
 * Lin (2014)... doesn't have an answer. Multiple underlying representations are possible, assuming optimality theory...
 * Lee & Zee (2003) write that the palatals can only ever be followed by "[i] or [y]". Again using square brackets.
 * According to Wu (1994), palatals are underlying velars that undergo palatalisation and have a pre-nuclear glide as part of the onset. Barefoot Banana (talk) 20:11, 10 September 2023 (UTC)

Tone sandhi / Second and fourth tone change / conflicting context
The section "Tone sandhi / Second and fourth tone change" uses the term "conflicting context". I had a look at reference [43] (Xu, Yi (1993). Contextual tonal variation in Mandarin Chinese) and I think I understand what "conflicting context" means. However at this part of the section I feel it's counterproductive to use it, as it's not a standard term; it confuses more than it helps. One can describe this sandhi by simply listing the cases in which it occurs; so it is possible to not mention "conflicting context".

So I suggest changing this somehow. Maybe it's best to go simply. How about this suggestion as an improvement?

"In conversational speech, for the rising tone (tone 2) and falling tone (tone 4), in some situations (depending on which tones are used immediately before and after), the pitch contours will change."

The wording is a bit clumsy and has room for improvement, but I think it's better than the current version which mentions "conflicting context" without explaining it.

Still, maybe it would also be good to explain the idea of "conflicting context". Perhaps a short explanation of conflicting context vs non-conflicting context afterwards may be helpful too. Or else a link to an article about it?

I do not feel qualified to do this second change (explanation of the term), but the first change I suggested above is quite simple, and I'm tempted to go ahead and do it. I may actually do that after a reasonable wait - maybe a week or so. GoPlayerJuggler (talk) 16:07, 17 April 2023 (UTC)

Description of alternative vowel analyses as 'Pinyin-based' and 'Bopomofo-based'
This seems odd. Phonological analyses of a language are not supposed to be motivated and justified by one or several transliteration systems that happen to be used for the language, but by phonological theories, analytical priorities and empirical facts about the language (as spoken). The analyses should be neither more nor less correct depending on how a language is spelt and whether it has a standard spelling at all. I'm sure that the linguists who proposed the analyses of Mandarin vowels in the article didn't motivate them as being 'based' on transliteration systems, because that would make them appear non-serious. This sounds more as if a Wikipedia editor has speculated that the linguists may have been influenced by these spelling systems, which would be original research.

As a matter of fact, neither of the analyses matches either spelling system very closely: the five vowel phonemes (/a/, /ə/, /i/, /y/, /u/) in the first analysis don't correspond 1-to-1 to the six vowel signs used in Pinyin (a, e, i, o, u, ü), and the two vowels (/a/, /ə/) in the second analysis don't correspond 1-to-1 to the vowel signs used in Bopomofo (three signs for full monophthongs ㄞ, ㄝ, ㄡ, three semivocalic or fully vocalic signs ㄧ, ㄨ, ㄩ, plus four unique diphthong signs ㄠ, ㄞ, ㄟ, ㄡ). The only parallelism I see is that Bopomofo spells semivowels and the corresponding full vowels in the same way (ㄧ, ㄨ, ㄩ), and they are indeed treated as underlyingly the same in the second analysis. However, that's hardly enough to claim that the analysis is 'based' on Bopomofo. 87.126.21.225 (talk) 21:32, 11 August 2023 (UTC)


 * Aye, it's odd. I think the distinction was grasping at the fact that pinyin uses an alphabet with (mostly) phonemic polygraphs, while bopomofo is more bespoke/specific to the traditional system of initials and finals. Remsense  留  16:45, 31 December 2023 (UTC)