Problems a Chinese girl has with writing one of the characters for her name
« previous post | next post »
My graduate seminar on Middle Vernacular Sinitic has six M.A. students from the PRC in it. They are all advanced in Literary Sinitic / Classical Chinese (LS/CC) . In this seminar, which I have been offering for more than a decade, each time I focus on a different medieval text. Because the texts I assign to the students are largely or wholly unannotated, the students are mostly sailing through uncharted waters. For them to be able to read and understand these texts, their Sinitic philological skills have to be high, higher than for most students in Advanced LS/CC courses.
By way of introduction, I requested the students to tell the class how to write their names in characters. For most of the students' names, with some back and forth explication, we could explain which characters were operative, although some were harder and some were easier. For the ones we couldn't readily explain verbally, we wrote them on the board. Of the 18 characters in question, there was one character — consisting of only 10 strokes (less than average) — that nobody in the room (other than the student to whom it belonged) could recognize. They couldn't figure out what it meant, how it sounded, or its construction. I had probably seen this character two or three times in my life, so I knew what it looked like, but didn't know much else about it. Just judging from its appearance, I guessed that it meant "a kind of (fine) jade" (there are numerous recherché characters for types of jade like that, as there are for horses, fish, usw.)
I asked the student who owns that character, Zhao Xutong, if I could write about it on Language Log, and she told me I could.
Xutong said:
At this point, I thought I was almost finished with this post, because — when I asked Xutong if it were in my favorite smallish dictionary, the famous Xīnhuá Zìdiǎn 新华字典 (The Xinhua Dictionary of Chinese Characters) — the world's best-selling reference work, with nearly half a billion copies in print, which I held up to the class, she said "yes", it was. But when I went home that evening and looked for it, try as I may couldn't find it, though I spent much time searching for it every which way.
珝 may be well nigh unrecognizable to most people, including highly literate and learned individuals, but it does pop up often enough (maybe once every several million times) that it does have to be taken into consideration by font designers and managers, government officials, financial institutions, professors, and so forth. It is not exceedingly rare in terms of the total corpus (approximately 100,000 sinographs), but it is quite rare enough to give plenty of people — including one person who has it in her name — lots of headaches.
This is not the first time I have encountered similar mystery / phantom characters in my classes or in public situations (see, for example, hàn 菡 below).
By the way, many years ago, perhaps before Language Log, I wrote that the first act of the Japanese Diet was for the members to explain how to pronounce their names. I don't know if that still obtains.
Technical notes on tóng 童
"child; servant boy; virgin; bare”
Löffler (1966) compares it to Kuki-Chin dong (“boy”); see also Rengmitca tong-kléng' (“boy”), Areng thon-dén (“boy”) (Löffler, 1960). Schuessler (2007) also compares it to Hmong-Mien: White Hmong tub (“son”), Iu Mien dorn (“son”).
“shaman”
Norman and Mei (1976) proposed that the Min Chinese word for “shaman” (*-dəŋA), written as 童, is from an Austroasiatic substratum, cognate with Vietnamese đồng, Mon ဒံၚ် (tòŋ, “to dance while under daemonic possession; to proceed by leaps”), ဒေါၚ် (tòŋ, “shaman called in to organise kəlok dances”). This is rebutted in Sagart (2008), who cited the wide distribution of the sense “magician; sorcerer” in late 19th-century & early 20th-century Chinese and the secondary meaning of 童 as “servant; messenger”, describing the resemblance between the Min and Austroasiatic terms as “undoubtedly fortuitous”.
Oracle bone form, an ideogrammic compound (會意 / 会意): 䇂 (“chisel”) + 見 (“kneeling person with a huge eye”). It depicts a person getting their eye (目) gouged out, a common punishment for slaves in ancient China. Compare 民, 臧.
When I began studying Chinese religions in the early 70s, one of the first terms I learned, one that captivated my imagination, was jītóng 乩童 ("spirit medium").
I was intrigued that a synonym in Hakka, Hokkien, and Teochew was 童乩 , with the syllables reversed.
Although the word was pronounced jītóng in MSM, I never for a moment thought that it had anything to do with northern religious practices (I witnessed Formosan tâng-ki climbing up ladders whose rungs were sharp knives and flagellating themselves till blood flowed copiously from their back.
Historical notes on Zhào 趙
Zhao (/dʒaʊ/; traditional Chinese: 趙; simplified Chinese: 赵; pinyin: Zhào; Wade–Giles: Chao⁴) is a Chinese-language surname.[note 1] The name is first in the Hundred Family Surnames – the traditional list of all Chinese surnames – because it was the emperor's surname of the Song dynasty (960–1279) when the list was compiled. The first line of the poem is 趙錢孫李 (Zhao, Qian, Sun, Li).
Zhao may be romanized as "Chiu" from the Cantonese pronunciation, and is romanized in Taiwan and Hong Kong as "Chao" in the Wade–Giles system. It is romanized as Vietnamese family name "Triệu" among the Chinese diaspora in Vietnam. Zhao is cognate to Korean family name "Cho" (조) in Korea.
Lexicographical notes on xǔ 珝
- Kangxi Dictionary: page 730, character 26
- Dai Kanwa Jiten: character 20952
- Dae Jaweon: page 1142, character 10
- Hanyu Da Zidian (first edition): volume 2, page 1114, character 7
- Unihan data for U+73DD
Since I couldn't find xǔ 珝 in the Xinhua Dictionary, which has about 10,000 characters, but it does occur in the above sources, which go up to 50,000 or more, you can get a sense of its frequency.
Selected readings
- "Recognizing half of a character and half of a word" (5/2/21) — on the character hàn 菡, which has a story similar to xǔ 珝
- "A confusion of languages and names" (7/18/16)
- "The political dangers of mispronunciation" (4/5/17)
- "How to pronounce the name of the ruler of the PRC" (10/16/25)
- "How to pronounce the surname 'Mair' and other Doggie talk" (2/17/22)
Stephen Goranson said,
March 6, 2026 @ 4:19 pm
This makes me wish to hear from her parents who named her.
Chris Button said,
March 6, 2026 @ 6:09 pm
I'm not following this line of reasoning.
Chris Button said,
March 6, 2026 @ 6:22 pm
I looked in the cited paper, and I'm not sure I follow the rejection of 札 in its sense of "die" as having an Austroasiatic connection either. Can't 殺 "kill", which is associated with Old Mon kcøt, also be tied in too?
David Moser said,
March 6, 2026 @ 10:51 pm
The character is searchable on my Pleco app (on which I've added about a dozen digital dictionaries) but it doesn't show up on the Chinese input on my MacBook. Very rare character indeed. This parental wickedness is seen in alternate spellings of common names: Crystal spelled as "Chrystle" or "Krystall", Abigail as "Abagayle" Jackson as "Jaxson" etc.
Peter Cyrus said,
March 7, 2026 @ 3:33 am
Can I ask a background question?
In an orthography that represents sound, more or less well, one needs to know the language to connect the written form with the meaning: [nuw] means different things in English and French. The written forms also differ – "new" and "nous" – because they're different orthographies, but their variation is constrained by the sound: we couldn't write "new" as "fgwx". The connection between sound and meaning is only etymological, and simply has to be memorized.
In Chinese orthography, the grapheme represents meaning, and the connection with sound is indirect via language: the character 月 means "moon" or "month" in several unrelated languages that pronounce it quite differently (yuè, tsuki, wel). Nonetheless, in many cases, characters are composed of a radical that hints at meaning and a phonetic element that hints at sound, e.g. 問 kinda sounds like 門 but has something to do with your mouth.
In the case of 珝, something different seems to be going on. The two parts – 玉 jade and 羽 feather – both seem to indicate meaning, and the similarity of the sounds of 珝 xǔ and 羽 yǔ is coincidental. Or is the sound of 珝 xǔ derived from 羽 yǔ?
So I’m imagining someone describing this particular kind of jade as “feather jade”, which would be 羽玉 yǔyù. Then someone else says “hey, we can put the jade character inside the feather character” – maybe that seems more elegant, or more official. Then where did xǔ come from? Corruption of yǔ?
Is 珝 just a rare case of a compound character representing its meaning directly? Or is this more common than I realized?
Victor Mair said,
March 7, 2026 @ 7:01 am
Confronting the issues raised by 珝 will help us get to the nitty-gritty of the sinographic writing system.
David Marjanović said,
March 7, 2026 @ 2:15 pm
The hints worked a lot better when the characters were created 2000–3000 years ago.
Victor Mair said,
March 7, 2026 @ 3:25 pm
"Is 珝 just a rare case of a compound character representing its meaning directly? "
It doesn't even do that. When I said that it might represent a type of fine jade, that was just a wild guess. Most people don't recognize which part is the radical and which the phonophore. If they guess that the radical is the part on the left, 王, they usually surmise incorrectly that it means "king", not "jade". And, as Zhao Xutong also informed us, they have no viable clue what it sounds like. The vast majority of people who look at this character are simply at sea.
When we look at "Chrystle", "Krystall", "Abagayle", "Jaxson", etc., we can make a reasoned stab at each of them, and it won't be far off. I can even make a reasonable approximation at Serbian Jokić, a name I never saw until about a month ago, but one that I'm seeing a lot in the last few weeks.
Chas Belov said,
March 7, 2026 @ 5:16 pm
Yes, I would have thought "king" although there doesn't seem to be a king radical. I see in my macOS symbol viewer that all the characters that use the jade radical have this property.
I was able to find 珝 at U+73DD.
Yves Rehbein said,
March 7, 2026 @ 11:52 pm
Then where did xǔ come from? Corruption of yǔ? I guess that may be it.
Following the previous thread about cinema I looked at some lemmas to do with images and wondered if xíng 形 is related to yǐng 影. From the reconstruction /*qraŋʔ/ respectively /*[ɢ]ˤeŋ/ it might seem fortuitous, though it need not be regular if there are loanwords in the mix, e.g. Austroasiatic in view of Schuessler: "Cognate with 英 (OC *qraŋ, “brilliant; flower”) or rather its Austroasiatic comparanda, including Proto-Bahnaric *ʔaːŋ (“bright (light)”) (Schuessler, 2007) [apud Wiktionary: 影]
The whole issue is difficult for me.
One line of inquiry begins with xiàng ("elephant; idea, image"). The ancinet etymology (cf. Wiktionary) makes it similar to grýps / griffon (cf. Adrienne Mayor, previously on LLOG). Polygenetic origins are likely. That alone would be a blog post but this is not my blog.
Another problem is fāng 方 in zhèngfāngxíng ("square") and běifāng 北方 ("north") and the by-sense of 方 ("method, law"), as Wiktionary speaks hypothetically (after Schuessler) "… to an Austroasiatic allofam which includes 狀 (OC *zraŋs, “form; appearance”). "Law; norm; standard" is a common extension of "form; shape".
At that point, píng 平 ("flat") is remarkable as a similar term in the lexical vicinity, I think?
And then JP eiga resembles emoji, which is well known to be from e ("image"). Yin and Yang may also be related.
Chris Button said,
March 8, 2026 @ 11:37 am
珝 makes it into the abridged Morohashi.
I'm surprised no-one tried to read it as 栩, which would actually have produced the correct reading too.
Chris Button said,
March 8, 2026 @ 11:44 am
翊 has caused headaches for some researchers. Baxter & Sagart seem to reconstruct it with a -p coda, but that seems unlikely given that it is used interchangeably with 羽 in the oracle-bone inscriptions.
Richard Futrell said,
March 9, 2026 @ 2:48 pm
Maybe you can help me identify a character I saw in a dream. It was 登 but on the inside, instead of 豆, was 官 or the inside of 追. I’ve never found it in a reference work. In the dream it signified ill portent.
Michael Watts said,
March 10, 2026 @ 6:00 am
I'm not sure what suggested that. The character refers to a type of jade, there is no connection to feathers, and you already stipulated that the sounds are similar. Why would you conclude, from the fact that 羽 lacks a semantic connection to the character but possesses a phonetic one, that it must be a semantophore and not a phonophore?
The 说文解字 says "从玉羽聲", much like you'd expect.
Tye S Power said,
March 10, 2026 @ 9:01 am
We named our daughter Eowyn. It is interesting to compare her experience explaining her name with the experience of Victor's student. Eowyn can cite the literary reference, which clears it up for some people, and then just spell her name. It is so much more complex for Victor's student. I am curious why she was given this name.
Jerry Packard said,
March 10, 2026 @ 2:19 pm
“ when I asked Xutong if it were in my favorite smallish dictionary, the famous Xīnhuá Zìdiǎn 新华字典 (The Xinhua Dictionary of Chinese Characters) — the world's best-selling reference work, with nearly half a billion copies in print, which I held up to the class, she said "yes", it was. But when I went home that evening and looked for it, try as I may couldn't find it, though I spent much time searching for it every which way”
It is in the most recent (12th) edition – on p. 546, ‘一种玉’ ("a kind of jade").
Jonathan Smith said,
March 12, 2026 @ 12:08 am
Generally "玉" + X names are explained wrt meaning as "kind of jade" and wrt sound ideally sound like X. Certainly people will assume they sound like X. If they don't sound like X, that's damn annoying, b/c they're generally not any real "kind of jade" you could ask for in say a jade shop. Better gloss would be "girl's name; visually evokes jade; not an actual kind of jade."
Yes Sagart re: "童" is one of the thoroughly wrongest article I've ever read. The author is deeply invested in Austronesian connected to Chinese, NOT Austroasiatic connected to Chinese. I don't know.
Interesting Things of the Week: 20/03/2026 | No Man's Guy said,
March 20, 2026 @ 4:29 pm
[…] Log: Abstand und ausban (2025), Problems a Chinese girl has with writing one of the characters for her name (2026), and Irish in Jamaica […]
KIRINPUTRA said,
April 1, 2026 @ 9:47 am
@ Victor
Whether the TÂNG- in Hoklo TÂNG-KI is etymologically coordinate to 童 or not has yet to be established, as far as I know. It seems very possible.
As far as the writing goes, the prevalent customary forms in Formosa are 銅乩 and 銅枝, where 銅 and 枝 are pure sound graphs — go-to sound graphs, in fact, for the respective syllables.
The 蔡俊明 Teochew-Mandarin dictionary has 同機, although 銅乩 and 同乩 pull up Teochew-related hits on the internet as well. The 蔡俊明 dictionary is a unicorn & a treasure — it’s a modern Hoklo dictionary that presents sinographic usage w/o swapping in pet hallucinations. It’s not pan-dialectal though — it’s (refreshingly) based on Teoyeo 潮陽 speech & usage.
https://cuhk.edu.hk/ics/clrc/chinese/pub_monographs_6.html
童乩 is a motivated scholarly hallucination or similar. It may be that 童 is customary in Hokchew (Foochow, 福州語) for the cognate of TÂNG-; in that case, 童〜 might’ve been an idle, if motivated, scholarly borrowing — the gentlemen of late antiquity didn’t write vernacular.
Point is, written usage for Hoklo can’t be truthfully known from Sinological writings (and certain other classes of writings) even when they appear to explicitly present written usage. They’re systematically off, with collective ideological intent — although any given individual may be completely naive. 蔡俊明 (for Teochew) and the 台字田 (for Taioanese) are useful for checking written usage, but it stands to reason that there’s no shame in using naked romaji when writing about Hoklo, just as it’s common to present Vietnamese words in romaji w/o their Hán Nôm.