Ad hoc Romanization for Mandarin: 2022 Winter Olympics

« previous post | next post »

"Beijing reveals mascots for 2022 Winter Games", YouTube (9/19/19):

Here are the names of the mascots of the 2022 Beijing Winter Olympics in official Hanyu Pinyin (HP):

Bīng Dūndūn 冰墩墩 (a roly-poly panda)

Xuě Róngróng 雪容融 (a lantern child)

So there's a question of why the organizers of the Beijing Olympics decided to use nonstandard spelling for these names.  They claim that "Bing Dundun" and "Xue Rongrong" are harder for foreigners to pronounce accurately (to sound something like Mandarin) than the ad hoc spellings they've come up with:  "Bing Dwen Dwen" and "Shuey Rhon Rhon".

I beg to differ.  I think that "Bing Dwen Dwen" and "Shuey Rhon Rhon" look ghastly (with three capitalized syllables for their "surname" and given name) and sound worse than HP "Bing Dundun" and "Xue Rongrong" in the mouths of foreigners.

Both versions have "Bing" for the surname of the panda, which is a freebie, since we have that syllable with the same sound in English (the name of the cherry and also of a very famous male singer).  Beyond that, though, it gets tough in both versions of both names.

People who don't know anything about Mandarin phonology are likely to say "done-done" for "Dundun".  Indeed, English has "dun" meaning "a dull grayish-brown color".  But "Dwen Dwen" is apt to come out sounding like "d-when d-when" for English speakers.  If you want to go ad hoc for "dundun", "doondoon" would work better.

While it's true that many Chinese do not distinguish between final -n and -ng, English speakers have no problem with that distinction (e.g., "ban" vs. "bang").  So there's no excuse for dropping the "-g" at the end of "Rhon".  I don't know what prompted them to do that, unless someone from a topolect that lacks the "-g" final was in charge of drawing up these casual Romanizations for the mascots.

As for the vowel of HP "rong", it is pronounced differently by individual foreign speakers, but the whole syllable is often heard as sounding like "wrong".  At least that's not an atrocity for "Rongrong".  But what can a foreign speaker make of "Rhon"?  Maybe something like the man's name "Ron" with a bit of breathiness added in after the "r".

Now, both for HP "Xue" and ad hoc "Shuey", we've got problems.  To tell the truth, the initial "X-" of "Xue" is one of the hardest three letters in HP for foreigners to pronounce (as epitomized in the name of the novelist Cao Xueqin 曹雪芹 [mid-18th c.]); they murder it in diverse ways that I won't even attempt to replicate here.

So much for the initial of "Xue".  How about the final, "-ue"?  Oddly enough, and this really perplexes me, even students who have been through 3 or 4 years of instruction in Mandarin often render it as "-way", and many foreign speakers of Mandarin never get over mispronouncing "xue" as "shway", both for xué 学 ("study") and xuě 雪 ("snow"), which are exceedingly common words, alas.  So I cannot understand why the Beijing organizers would steer foreign speakers to mispronounce "xue" as "shuey" from the get-go.

All in all (except for the "x" at the beginning of "Xue"), HP is easier for foreigners to pronounce "Bīng Dūndūn 冰墩墩" and "Xuě Róngróng 雪容" to sound roughly like Mandarin than it is for them to wrestle with "Bing Dwen Dwen" and "Shuey Rhon Rhon".

It is surpassingly strange that, after energetically promoting HP for more than half a century and making it nearly universal within China, even convincing the UN and ISO to accept it as the standard Romanization for Modern Standard Mandarin (MSM) and getting most nations on board with it, they would now backtrack to something that is in no way superior to HP.  That way lies senseless chaos.

It's different with "Loose Romanization for Cantonese" (9/21/19), since Cantonese, like all Sinitic topolects, has never had a standard Romanization in widespread use.  Furthermore, the "Loose Romanization" that is coalescing among the anti-extradition, pro-democracy protesters in Hong Kong these days is developing in an organic fashion, being created for practical purposes by the users themselves.

A few words about the meanings of "Bīng Dūndūn 冰墩墩" and "Xuě Róngróng 雪容融".  Using dictionary definitions, one could refer to them as "Chubby / Stumpy / Short-and-Stout / Squat / Blockish / Heavy-set / Pier-like Ice" and "Tolerant / Accommodating-Melt / Thaw / Blend / Merge Snow".

Here are some comments from PRC native speakers and an advanced foreign speaker (no. 4):

1. Reduplicatives sound cute….  From my perspective, 墩墩 is especially very cute and related to winter. For example, little kids wearing thick clothes are "胖墩墩", anything frozen into ice is "冰墩墩". My feeling is that the image particularly matches my memory of winter in northern China — ice and snow almost everywhere, and little chubby kids running around together everywhere.

2. Those reduplicatives remind me of Shaanxi topolect.  We don't usually use that type of format of words in Putonghua, except when talking to babies. 冰墩儿 is  more common  than 冰墩墩 in Putonghua.

3. "墩儿" is a common term in North China, often referring to someone who is healthy, strong, outgoing, and so on, which is quite suitable as a description for children. However, the people responsible for making up the mascot names must have realized that 儿 (retroflex "r") is difficult for people living in southern China as well as for foreigners to pronounce (I cannot really pronounce it myself), so they changed it to "Dundun" instead of "Dunr".

4. When reading Pinyin, Americans would tend to pronounce the "x" in "xue" with a "zh"-like sound ("zhuey"), as opposed to something more like a "sh" sound that is closer to its Chinese pronunciation.

5. the two names are representative of Chinese people's fondness for injecting as much homophone information as they can into names.

Thus 墩 sounds like 敦, meaning 敦厚,敦实, firm, grounded. The reduplicated syllables 墩墩 deliver a childish cuteness.

容融, with 容 meaning 容纳,acceptance, containment, and 融 meaning 融合, fusion, convergence, coalescence.

I guess only Chinese people can get the meanings of the names, though the organizers must have wanted them to be understood by people all over the world. I personally don't like spacecrafts being named curiosity, or ambition, which are too hegemonic as values.  However, they are, indeed, easy to understand and memorize. The dissemination-friendly quality, I think, outweighs the cryptic intimations contained by 冰墩墩 and 雪容融.

6. To tell you a secret: I have a nickname in my family, which is……墩墩!!!!!!!! Because I was a very chubby baby, when I was apparently not old enough to put in a word for my own nickname. [P.S.:  She still has this nickname, even though she is now tall and slender!]

So as you can imagine…I felt very…cold?… when I saw the name 冰墩墩 :D

I would say 墩/墩墩 is fairly common in China, especially in northern China, to lovingly call a chubby baby or a kid, since people always associate the character 墩 with the word 胖墩墩.

In speaking of the mascot of the 2022 Beijing Winter Olympics 冰墩墩, the organizing committee interprets its name as follows: "冰,象征纯洁、坚强,是冬奥会的特点。墩墩,意喻敦厚、健康、活泼、可爱,契合熊猫的整体形象,象征着冬奥会运动员强壮的身体、坚韧的意志和鼓舞人心的奥林匹克精神。" (from this Baidu encyclopedia entry) According to this explanation, "墩" here refers to honesty, robustness, and fitness. 雪容融, the name of the mascot of the Paralympics, is officially interpreted as "雪,象征洁白、美丽,是冰雪运动的特点。容,寓意包容、宽容、交流互鉴。融,寓意融合、温暖,相知相融。容融,表达了世界文明交流互鉴、和谐发展的理念,体现了通过残奥运动创造一个更加包容的世界和构建人类命运共同体的美好愿景。" Basically, "容" refers to inclusiveness , while "融" represents fusion and harmony. I think it intends to indicate the spirit of the Paralympics.

7. I watched the video, and it is interesting that it explained "墩墩" as "robust and lively." In my opinion, the sense of "墩墩" is closer to "fat" or "plump," particularly in a cute way, because of the expression "胖墩墩." And the panda mascot has a similar feeling with round face and body, which I don't think it is quite fit to the expression of "robust and lively." But I still prefer the name "冰墩墩" because it is easier to pronounce for people who come from the southern part of China. I don't like 翘舌音 (retroflex sound).

As for the usage of romanization, I think it might relate more to politics. President Xi emphasizes soft power (文化强国 [VHM: "cultural power / powerful nation]) and China has been working on the output of culture for a long time. One obstacle is that Pinyin is hard to pronounce for people who do not know it before, and thus ad hoc Romanization would be a good solution. And since the Olympics is such a huge event, the government does not want to miss this opportunity. I think this is a good phenomenon, an effective way for promoting culture which requires not only indigenous essence, but also compromise to globalization. Romanization is just the first step.

8. I think the names are cute and suitable for these mascots. As 冰墩墩 is a panda, "墩墩" means it is cute and robust. Chinese people like to use the word "胖墩墩" to describe chubby babies. And "雪容融" is a red lantern (灯笼); the name means that snow can be melted (by the lantern). 容: allow, let, likely/easily; used as passive. 融: melt.

There is a Chinese proverb, "正月十五雪打灯" ("on the 15th of the first month, snow covers the lanterns"), which predicts the following year will be auspicious. I think the good connotation in this proverb is included in the images and names of these mascots.

9. I just came up with an idea that 容融 can also refer to 容易融化(melt easily). Since PRC is standing on a kind of awkward position in international relations, maybe the officials hope the 雪 (snow or ice) can be melted easily by holding the 2022 Winter Olympics.

I'm surprised to find out that PRC is using ad hoc romanizations rather than Hanyu Pinyin. Very similar with the Hong Kong way. I feel there might be a change in the direction of the political winds. Maybe they want to pay more attention to international attitudes?

10. Oh!!!!! After staying away from the news for a while, it appalls me how terrible these two names are! To my view, 冰墩墩 and 雪容融 are two of the most hideous names I have ever seen given to any mascots in the history of Chinese events. As a Chinese myself, my patriotic neurons viscerally reject these two names, if one calls them "names". Who let these names pass and enter the final round? 關係戶嗎? [VHM: somebody with connections?]

So pedantic and artsy-fartsy! One can envision a greasy official, who hardly ever knew any Chinese Classics but their titles, turning the pages of a character-dictionary trying to find something quaint, but ended up neither fish nor fowl. These names remind me of those of the helpless first-grade kids full of awkward pedantry that their beloved parents thought would make sense. Their namer does try hard to sound learned, one can tell.

I have nothing to say about 冰墩墩 but envisaging a big solid chunk of ice. And nothing else can be inferred aside from its literal sense. 容融 indeed has some positive nuances. 容 means "embracing, containing" or "to permit, authorize", and 融 "to melting" or "to intermingle". So, 雪容融 literally means "snow-permits-melting". So perhaps this name signifies how China will allow the existence of a melting-pot-like multiculture (well, has it not, ever, for millennia)? Or, China still welcomes those who take (or are taking) a different form (even if you melt into a liquid state, we solid-state snow still embrace you)….? Is this name connotative of what's happening in Hong Kong right now, an awkward claim and cry that 你還是我們的同胞 ("You're still our compatriots") whatever beliefs you hold? But who are you to "authorize/permit/容" the existence of a belief, or a belief-holder's affiliation?

In any case, it baffles me why in the world these two names win out. I miss the simple, 2008 Beijing Olympic mascot names.

As for Bing Dwen Dwen and Shuey Rhon Rhon, they only add to my detestation. The sole reason that instantly comes to my mind justifying why pinyin spellings are not used is the Cantonese romanization in the Hong Kong protest, about which you just wrote in your last LL post. Is the government trying to cater to the protestors' behavior as a gesture of reconciliation? But this "Bing Dwen Dwen – Shuey Rhon Rhon" romanization is nowhere close to the way that Cantonese speakers romanize their language, or even to any romanization systems that existed in history (viz., Gwoyeu Romatzyh or Wade-Giles)!

Phonetically it's also bad. It cannot represent the Mandarin pronunciation well. For 冰墩墩 /pɪŋ tun tun/, the /u/ in the middle is slightly unrounded. "Dwen" with a [w] medial is misleading. But this is nothing as excessive as 雪容融. For "xue rong rong", /ɕye ɹoŋ ɹoŋ/, romanization the "snow" (pinyin: xue) as "shuey" is ignorant. Since /ʂ/ and /ɕ/ are in complementary distribution in Mandarin, any "sh" before a back vowel such as /u/ would be intuitively read as /ʂ/ instead of /ɕ/ even by native speakers. This complementary distribution of /ʂ/ and /ɕ/ is the exact reason why many native speakers of Mandarin cannot pronounce the 3rd person female demonstrative "she" and utter a "xi" (like in Chairman Xi) sound instead. — So, "shuey"??? At least, for the sake of Mandarin phonetic distribution, romanize something like "shyue" for "xue" (snow), so that "sh" precedes a front vowel! Whoever the romanizer is does NOT have the slightest sense of how Chinese phonology really works — or bother to learn.

Thumbs down in a nutshell and I feel disturbed.

All things considered, "Bing Dwen Dwen" and "Shuey Rhon Rhon" are a flop, if not an outright disaster.



"Diminutives and reduplicatives in Chinese " (12/31/15)

"Just what they DIDN'T want " (9/22/08)

"China's 2022 Olympic mascots have unusual names", by Qin Chen, inkstone (9/18/19; updated 9/19/19)


[h.t. Mark Swofford; thanks to Lin Zhang, Tong Wang, Nick Tursi, Diana S. Zhang, Yunzhu Huang, Chenxi Ouyang, Qing Liao, Yijie Zhang, Yishu Ma, and Chenfeng Wang]


  1. Neil Kubler said,

    September 30, 2019 @ 10:34 am

    I agree that despite the good intentions, it would have been far better to remain consistent and employ standard Pinyin to write Bing Dundun and Xue Rongrong. For the reference of foreign broadcasters, one could have perhaps added a simple pronunciation guide after the Pinyin, for example: Bing Dundun (pronounced Bing Dwoondwoon) and Xue Rongrong (pronounced Shuey Wrongwrong). [And I bet the main reason for dropping the final -g in "Rhon Rhon" is that it would then rhyme with the inauspicious English word "wrong"; it's true that hundreds of millions of Chinese speakers confuse final consonants -n and -ng, but this is usually in finals -in/-ing and -en/-eng, not in -ong]. As Bernhard Karlgren wrote many years ago, no romanization for Chinese is ideal for fulfilling all possible roles (use by native speakers, pronunciation guide for foreign speakers, etc.). If China truly wants Chinese to become an international language, it should insist on using only Pinyin (and I'll choose to retain the obvious ambiguity of that statement!).

  2. Ellen K. said,

    September 30, 2019 @ 11:04 am

    Though it's not clear to me how the "-ue" of Xue should be pronounced, "shuey" seems like a very odd romanization or pronunciation guide because I would assume "Xue" is a single sylable, but, for me, "shuey" indicates a two syllable word, "SHOO-ee", rhymes with chewy.

  3. Filipe said,

    September 30, 2019 @ 11:37 am

    I believe Bing Dwundwun and Syue Rungrung, as in standard Yale romanization, are the most intuitive spellings for English speakers. (Ok, maybe "roong" would be better than "rung", but I don't like to stray from the standard).

    And they could also use Ping Touen-touen and Siue Jong-jong as alternative spellings for those who speak French, EFEO-style.

  4. Philip Taylor said,

    September 30, 2019 @ 1:25 pm

    Listening to Google Translate's audio rendering of 冰墩墩, I hear the vowel of the first 墩 as being close to Wells' NURSE vowel, whilst the vowel of the second 墩 as more of a GOOSE-DRESS glide. Do others hear this, and if so, what is the most probably reason for this difference ?

  5. David L said,

    September 30, 2019 @ 2:14 pm

    I don't see how Xue is at all helpful to the English speaker who knows nothing of Chinese (like me). I have no idea how I am supposed to pronounce it. "Shuey," as Ellen K. says, looks like shoo-ey. Filipe suggests 'Syue' which I can only guess is something like the name Sue with a 'y' in the middle. Is that closer?

    I don't know what I'm supposed to make of Rhon versus Ron…

  6. Filipe said,

    September 30, 2019 @ 2:54 pm

    To David L:
    Take an s;
    Add to it the vowel in French rue "street";
    Then add to those the vowel in French bœuf "beef".

    The result should be like French sueur "sweat", but without the final -r.

    (The initial sound of xue is not _exactly_ s-, but it's close enough.)

  7. David L said,

    September 30, 2019 @ 3:15 pm

    Filipe: thank you, that's a nice explanation, and luckily I know enough French to be able to make sense of it. But the difficulty of conveying the sound of 'xue' only reinforces the impossibility of coming up with any simple transcription that will work for English-speakers without a good deal of further elaboration.

  8. Victor Mair said,

    September 30, 2019 @ 4:27 pm

    Yale is sywe and Wade-Giles is hsüeh. Do either of them work for you?

    How about shyueh?

  9. John Rohsenow said,

    September 30, 2019 @ 4:40 pm

    "All in all (except for the "x" at the beginning of "Xue")"
    "Wade-Giles is hsüeh:
    Yes, I always found that for some reason hs- seems to work for x-
    and "Hsueh" pretty close for xue. I think those old guys put some thought into those. :-)

  10. Victor Mair said,

    September 30, 2019 @ 4:58 pm

    Yes, John.

    I've always found Wade-Giles to be remarkably close to IPA in its principles.

  11. Jenny Chu said,

    September 30, 2019 @ 5:23 pm

    Speaking not from a linguistics point of view but from a branding point of view – trying to get people to pronounce anything the way you want it can never be done only through spelling. People are perverse and will ALWAYS find a way to say it weirdly. Instead: audio advertising, a jingle, and maybe a fun mnemonic ("remember – it rhymes with XXX!") would be more effective. Anyway, people of all nations, including China, are terrible at predicting how "foreigners" will behave.

  12. Filipe said,

    September 30, 2019 @ 6:12 pm

    Tsk!, that's right: Yale is sywe, not syue. I often forget it, probably because of Tongyong Pinyin influence. Sorry!

  13. Andrew Usher said,

    September 30, 2019 @ 6:22 pm

    I almost hate to point this out, but none of the text here so far gives the actual pronunciations (or the closest one could get with English phonology). Nor any enlighenment on why they would chosen 'Dwen' if the vowel is /u/.

    k_over_hbarc at

  14. David Marjanović said,

    September 30, 2019 @ 6:40 pm

    So I cannot understand why the Beijing organizers would steer foreign speakers to mispronounce "xue" as "shuey" from the get-go.

    It's really difficult to explain ü in writing to people who aren't already used to it ([y ~ ɥ]). Apparently, English monoglots have a strong tendency to interpret it as [ju]; pronounce you very slowly, and an [y] just might appear sometime in the first third of the word.

    Nor any enlighenment on why they would chosen 'Dwen' if the vowel is /u/.

    That's because the vowel isn't [u], it's a diphthong – [uə̯] or [uɤ̯] or some such.

  15. Twill said,

    October 1, 2019 @ 12:05 am

    The only explanation for "Dwen" I can imagine is that a) the transliterator cannot produce the sequence /we/, and thus vocalizes the glide, and b) for some mad reason decided that their resultant pronunciation better represents Mandarin's /u/ than the sensible correspondence /u:/. I don't really fault as an effort at representing something very alien to English phonology and at least the rest have some rhyme and reason to them (as stupid as they may be).

  16. Twill said,

    October 1, 2019 @ 12:07 am

    Fault "shuey" that is.

  17. Chris Button said,

    October 1, 2019 @ 2:03 am

    That's because the vowel isn't [u], it's a diphthong – [uə̯] or [uɤ̯] or some such

    I think I'd favor going with a more bopomofo-style analysis of "dwen" (dun) as /tʊᵊn/ with the superscript schwa signifying a quasi syllabic coda in a similar manner to how the word "near" is /nɪᵊr/ in GenAm.

    I would assume "Xue" is a single sylable, but, for me, "shuey" indicates a two syllable word, "SHOO-ee", rhymes with chewy.

    It's weird they didn't at least go with a "w" as in "shwey". That works pretty well for the Shwedagon pagoda in Burma (granted Burmese "shwe" and Mandarin "xue" don't really sound much alike in reality)

  18. Michael Watts said,

    October 1, 2019 @ 2:26 am

    why they would [choose] 'Dwen' if the vowel is /u/.

    The only explanation for "Dwen" I can imagine…

    I don't really fault as an effort at representing something very alien to English phonology

    "xue" is quite alien to English phonology, but "dun" is not. To the English ear, it's just the sequence dw?n, where the ? represents either the FOOT vowel or the commA (schwa/reduced) vowel. Neither option is at all difficult to pronounce. The vowel /u/ is not involved.

    English speakers are more likely to read "dwen" using the DRESS vowel. That is wrong, but it's not like there was a better option — FOOT and schwa both lack any standard spelling.

    On the other hand, it's quite easy to imagine why the Chinese thought "dwen" would be clearer — the pinyin syllable "dun" is a perfect rhyme for the pinyin syllable "wen", but with a d- onset instead of (as it's traditionally viewed) a zero onset. "Dwen" just spells out the whole rime.

    For reasons that are not clear to me, but that were obviously bad, the pinyin finals "wen" and "wei" are "simplified" to "-un" and "-ui" where the initial is nonzero. (Similarly, "you" becomes "-iu".) Compare "Feng shui", which also illustrates the quality of the letter "e" the Chinese have in mind.

  19. Michael Watts said,

    October 1, 2019 @ 2:34 am

    (For those curious about the difference between pinyin "x" and an /s/, /s/ is a fricative produced by using the tip of your tongue to create a restriction in airflow near your alveolar ridge. "x" is a fricative produced by using the middle of your tongue to create a restriction in airflow near where your hard palate lifts up toward your nose.)

  20. Michael Watts said,

    October 1, 2019 @ 2:35 am

    (And "sh" is a fricative produced in the same place as "x", but using the tip of your tongue rather than the middle.)

  21. Keith said,

    October 1, 2019 @ 3:08 am


    A "very famous male singer"?

    Are you dreaming of a white Christmas and Winter Olympics"?

  22. David Marjanović said,

    October 1, 2019 @ 6:11 am

    I think I'd favor going with a more bopomofo-style analysis of "dwen" (dun) as /tʊᵊn/ with the superscript schwa signifying a quasi syllabic coda in a similar manner to how the word "near" is /nɪᵊr/ in GenAm.

    I'm fine with this analysis. I used brackets, not slashes, to indicate actual sounds rather than any analysis – because the actual sounds are what the English-speaking hearer will hear (and will then interpret without knowledge of Mandarin phonology).

    It's weird they didn't at least go with a "w" as in "shwey".

    That would make me think of shui rather than xue.

    But I have an [y] in my native language, and that may make a big difference as mentioned above…

  23. Chris Button said,

    October 1, 2019 @ 6:34 am

    To be clear, I think "dun" may well be pronounced with a syllabic nasal coda in the same way as "near" effectively has a syllabic rhotic coda in GenAm. I suppose a Japanese moraic "n" might be a good comparison. 'Tis but a hunch though…

  24. Ben said,

    October 1, 2019 @ 11:37 pm

    Inkstone bizarrely explained the situation thus:
    "The romanized spelling of the mascot names is loosely, but not entirely, based on Gwoyeu Romatzyh, a system originally developed by Chinese linguists in the late 1920s but is now rarely used."

    Doesn't seem particularly close:
    Xuě Róngróng (py) Sheue Rongrong (gr) Shuey Rhon Rhon (Olympics)
    Bīng Dūndūn (py) Bing Duenduen (gr) Bing Dwen Dwen (Olympics)

    Changing the tones on the names does get us a little closer.
    Xué Rōngrōng (py) Shyue Rhongrhong (gr) Shuey Rhon Rhon (Olympics)
    Bīng Dúndún (py) Bing Dwendwen (gr) Bing Dwen Dwen (Olympics)

  25. Thomas J. said,

    October 1, 2019 @ 11:47 pm

    I think Pres. Xi has been in the news enough recently that people many people have picked up on Pinyin ≈ English [ʃ]. If they missed that, though, wouldn't the government be interested in educating people of this fact? If the goal is improving soft power, wouldn't Beijing want to use the opportunity to introduce the world to its official Romanization scheme?
    Maybe in five years we'll be writing "Pres. Shiy" instead.

  26. Andrew Usher said,

    October 2, 2019 @ 6:53 pm

    Since it appears people disagree I'm not going to understand without hearing it just how the actual syllables in question are supposed to be sounded.

    Yes, it would be a good thing for international intelligibility if the stranger conventions of Pinyin were changed. But that would sacrifice uniqueness; the designers of Pinyin wanted to distinguish every Chinese phoneme even if the Roman alphabet didn't have room for it. You just can't map three voiceless sibilants onto 's' and 'sh', and the third isn't obvious.

  27. Michael Watts said,

    October 3, 2019 @ 2:36 am

    the designers of Pinyin wanted to distinguish every Chinese phoneme even if the Roman alphabet didn't have room for it. You just can't map three voiceless sibilants onto 's' and 'sh', and the third isn't obvious.

    The case that Mandarin "s", "x", and "sh" are three distinct phonemes is far from airtight. There is no phonological context that permits both "x" and "sh" — if you know the following vowel, then you know which of "x" or "sh" occurred. You could represent "x" as "sh" with zero loss of information.

  28. Akito said,

    October 3, 2019 @ 9:58 pm

    You could represent "x" as "sh" with zero loss of information.

    Oh, no. That would make "xi" (系) and "shi" (是) homophonous. I think "si"-"xi"-"shi" representing [sɨ]-[ɕi]-[ʂɨ˞] respectively are quite reasonable. An alternative might be "si"-"syi"-"sri".

  29. Michael Watts said,

    October 3, 2019 @ 11:53 pm

    Well, it's true that the vowels in "xi" and "shi" are represented the same way, and that the Chinese traditionally think of them as being the same thing, though they don't think of them as being phonemes — the traditional analysis is that the syllables "xi" and "shi" have initials but lack finals.

    That is obviously wrong — "xi" has the same final as "mi" / "ni" / "li", which is not traditionally thought of as a zero final in those other syllables — but it's true that there is also no context that permits both vowels, so the same argument I made against considering "x" and "sh" to be distinct applies here.

    But that doesn't prove that you can't represent "x" as "sh" while suffering no loss of information. The Wade-Giles system does that very thing, not for the "x"/"sh" pair itself (which is distinguished as "hs"/"sh"), but for the related affricates "j"/"zh" and "q"/"ch". Pinyin's j and zh are both ch in Wade-Giles, just as q and ch are both ch'.

  30. Philip Taylor said,

    October 4, 2019 @ 4:23 am

    Michael Watts' immediately preceding contribution led me to search for some background information, and I came across (for the first time) the Wikipedia article on Pinyin. I repeat just a small fragment here, but the entire article is well worth reading, IMHO, if one is not already familiar with the details.:

    Native speakers of English will decode pinyin spellings to fairly close approximations of Mandarin except in the case of certain speech sounds that are not ordinarily produced by most native speakers of English: j [tɕ], q [tɕʰ], x [ɕ], z [ts], c [tsʰ], zh [ʈʂ], ch [ʈʂʰ], sh [ʂ], h [x], and r [ɻ] exhibiting the greatest discrepancies.

  31. Michael Watts said,

    October 4, 2019 @ 11:17 am

    except in the case of certain speech sounds that are not ordinarily produced by most native speakers of English

    I would argue that the sounds of pinyin "sh" / "ch" / "zh" / "z" are indeed commonly produced by native speakers of English, and for "sh" and "ch" the spelling is shared. ([ɻ] is a more complex case — that sound is normal in English, but pinyin "r" represents a continuum of sounds and is pretty likely to confuse English speakers.)

    There's a big difference between the claims that "[ts] is not ordinarily produced by most native speakers of English" [flagrantly false] and "[ts] is, in English, not ordinarily spelled with the letter Z" [definitely correct, but not what the wikipedia article actually says].

  32. Ellen K. said,

    October 4, 2019 @ 2:16 pm

    Yes, [ts] exists in English, but not word initial.

  33. M. Paul Shore said,

    October 4, 2019 @ 3:38 pm

    It came out on a Tuesday* and my heart just fell,
    Shu-ey Rhon Rhon Rhon, Shu-ey Rhon Rhon.
    Why choose this cockamamie way to spell?
    Shu-ey Rhon Rhon Rhon, Shuey Rhon Rhon.
    It's an awful sight.
    It won't get read right
    By all the Anglophones:
    It's way wrong, wrong, wrong, it's way wrong, wrong.

    *The official unveiling of the mascots was on Tuesday, September 17th.

  34. Philip Taylor said,

    October 5, 2019 @ 5:30 am

    Ellen — "Tsetse fly". Fully integrated into the English language for some long time. "Tsundere" is a more recent adoption.

  35. Akito said,

    October 5, 2019 @ 6:30 am

    But that doesn't prove that you can't represent "x" as "sh" while suffering no loss of information.

    If you represent Pinyin "xi" as "shi", you'd be forced to represent Pinyin "shi" differently, as does Wade-Giles ("shih").

  36. Ellen K. said,

    October 6, 2019 @ 7:58 am

    Okay, so what I said isn't true in an absolute sense, but it's true as a generalization, which is what's significant.

  37. Philip Anderson said,

    October 11, 2019 @ 7:12 am

    @Ellen K, Philip Taylor
    Surely what matters is whether an initial 'ts' sound is familiar to English speakers? I don't think 'tsetse' is that familiar, and it is often pronounced with an initial 't' anyway. 'Tsar' is common, but normally pronounced with initial 'z'.

    However, 'tsunami' is likely to come to mind, at least in Britain where I hear 'ts' from newsreaders ('s' as well). 'Tzatziki' too.

RSS feed for comments on this post