Language Log

« previous post | next post »

We've been looking at strange Chinese characters:

"Really weird sinographs" (5/10/18)

"Really weird sinographs, part 2" (5/11/18)

For a sinograph to be weird, it doesn't need to have 30, 40, 50, or more strokes.  In fact, characters with such large numbers of strokes might be quite normal and regular in terms of their construction.  What makes a character bizarre is when its parts are thrown together in unexpected ways.  On the other hand, characters with only a very small number of strokes might be quite odd.  Two of my favorites are the pair 孑孓, which are pronounced jiéjué in Modern Standard Mandarin and together mean "w(r)iggler; mosquito larva".

When I first encountered 孑孓 half a century ago, I burst out laughing.  Both halves looked as though they were the very common three-stroke character zǐ 子 ("child; son; master; seed"), but written by a drunkard who couldn't get the arms of the child straight, with jié 孑 sagging down on one side and jué 孓 sagging down on the other side.

Both 孑 and 孓 are classified under Kangxi radical #39, and both are said to have zero additional strokes.  Under the same radical and also with zero strokes, there is 孒, complete with its own Unicode number.  Guess what?  It's pronounced jué and allegedly means "mosquito larva", so it's the same thing as 孓, yet different.  Who needs it?  Furthermore, I doubt that it has ever in history been used in a sentence by itself to mean "mosquito larva".  (I'll explain more about that in a moment.)

If there's a pathetic, useless 孒 for quaint 孓, surely 孑 ought to have a corresponding partner.  Well, it does — sort of, but not quite.  There's (U+2193C), which you might not be able to see in your browser.  It looks just like 孑, except that the slightly upward slanting horizontal stroke doesn't cross through the vertical axis.  Instead, it just stops dead when it hits the vertical axis, so it's not exactly correspondent to 孒, where the horizontal stroke on the right side also stops dead when it hits the vertical axis, but is level.  Nonetheless, we may say that (U+2193C) for 孑 is analogous to 孒 for 孓 without being exactly correspondent.  (N.B.:  As it appears in Unihan, the horizontal stroke of (U+2193C) is level, but elsewhere [as in zdic], it slants upward.)

I asked several highly literate native speakers of Mandarin if they were familiar with the characters 孑 and 孓, their pronunciations jié and jué, their meanings separately and together, as well as the meaning of the spoken Mandarin word "jiéjué", and drew a blank on all of these.  When I asked them how they would say "mosquito larva" in Mandarin, they said something like "wénzi yòuchóng 蚊子幼蟲", which means exactly that:  "mosquito larva".  When I told them that's what jiéjué 孑孓 means, they said that they think it must be an ancient, arcane term.

It's not surprising that literate readers of Modern Chinese texts, even those who have a relatively high degree of proficiency in Literary Sinitic / Classical Chinese, would not be familiar with jiéjué 孑孓 since 孑 is ranked no. 4691 in a frequency list of 9,933 "common" characters (total number of characters in the corpus: 193,504,018), while 孓 is no. 6370.  In a further discussion of the shapes, meanings, and usages of these two characters below, we will see why 孑 has a much higher ranking than 孓.

The oldest occurrence of jiéjué 孑孓 that I can find is in an annotation by the Eastern Jin period scholar Guo Pu 郭璞 (276-324) on the Ěryǎ 爾雅 (Approaching Elegance) (3rd c. BC), the earliest surviving lexicon (perhaps more accurately referred to as a synonymicon) for Chinese, which is especially strong on terms for flora and fauna.  Guo Pu's commentaries are a gold mine of local and vernacular terms in the living language of his time.  Jiéjué 孑孓 is undoubtedly one such non-literary term, which would account for its minimal transmission across nearly two millennia to the present day.

My intuition is that, like many early disyllabic terms referring to animals, especially insects, the jié 孑and jué 孓 of 孑孓 jiéjué originally did not mean anything individually.  In other words, I think that 孑孓 was a disyllabic morpheme.

I further surmise that 孑孓 may be the dimidiation of a single word with complex phonology (consonant cluster, etc.), and that it may have a non-Sinitic origin.

Finally, I suspect that the sinographs 孑 and 孓 were devised via deformation of zǐ 子 ("child; son; seed").

What might the original word lying behind Mandarin jiéjué 孑孓 have sounded like?

The Cantonese pronunciation of these two sinographs would be kit3 kyut3.  The Middle Sinitic reconstruction by Zhengzhang Shangfang is /kˠiᴇt̚  kɨoŋ^X/, and his Old Sinitic reconstruction is /*ked  kʷoŋʔ|kʷad/.  It may be noted that the expected MSM reflex of Middle Sinitic kɨoŋ^X would be jiǒng, yielding jiéjiǒng instead of the jiéjué in current use.  (Source of reconstructions)

Jonathan Smith suggests an association with the Minnan writing 車蛆 for "mosquito larva".  The two characters literally mean "car maggot".  The first character is evidently being used transcriptionally.

• (Hokkien: Quanzhou)
  • Pe̍h-ōe-jī: chhia-chhir
  • Tâi-lô: tshia-tshir
  • IPA (Quanzhou): /t͡ɕʰia³³ t͡sʰɯ³³/
• (Hokkien: Xiamen)
  • Pe̍h-ōe-jī: chhia-chhu
  • Tâi-lô: tshia-tshu
  • Phofsit Daibuun: chiazhw
  • IPA (Xiamen): /t͡ɕʰia⁴⁴⁻²² t͡sʰu⁴⁴/
• (Hokkien: Zhangzhou, Jinjiang)
  • Pe̍h-ōe-jī: chhia-chhi
  • Tâi-lô: tshia-tshi
  • Phofsit Daibuun: chiachy
  • IPA (Jinjiang): /t͡ɕʰia³³ t͡ɕʰi³³/
  • IPA (Zhangzhou): /t͡ɕʰia⁴⁴⁻²² t͡ɕʰi⁴⁴/

Source

I welcome suggestions from readers about how the Middle Sinitic and Old Sinitic reconstructions might have sounded undimidiatedly and how they may relate to the Minnan pronunciations.

Incidentally, the Cantonese, Minnan, Middle Sinitic, and Old Sinitic for zǐ 子 ("child; son; seed; master") are respectively zi2, chú / chír / chí, /t͡sɨX/, /*[ts]əʔ/, /*tsəʔ/ [Baxter-Sagart]

Now on to a brief investigation of what 孑 and 孓 may have meant, if anything, outside of their incorporation in the lexeme jiéjué 孑孓 ("w[r]iggler; mosquito larva").

It may seem curious that jié 孑 is listed in Paul Kroll's A Student's Dictionary of Classical and Medieval Chinese (Leiden:  Brill, 2015), p. 208a, but jué 孓 is not. Not only is jié 孑 listed, it receives elaborate treatment:

1. by oneself, singly, alone.

    a. rdup., sole and solitary; also, unique and unmatched, standing out alone; (med.) also, exacting, fastidious and finicky.

2. leftover, remainder.

3. ☉ 戟 Jǐ 1, guisarme, halberd.

The Shuōwén Jiězì 說文解字 (Explanation of Characters) (100 AD), the Bible of traditional Chinese character studies, states that jié 孑 means "lacking a right arm" and jué 孓 means "lacking a left arm".  I'm dubious.  Such overly convenient definitions are all too common in the Shuōwén and in other early lexicographical works for disyllabic morphemes, e.g., qílín 麒麟 ("kirin" — allegedly the male and female of this mythical species [I am going to write a major post on this term later on]), fènghuáng 鳳凰 ("phoenix" — also allegedly the male and female of this mythical species), pípá 琵琶 ("lute; al-'ud; barbat" — supposedly the upstroke and downstroke used in playing this stringed instrument).

Hànyǔ dà cídiǎn 漢語大詞典 (The Unabridged Dictionary of Sinitic), 4.175b-177a gives 17 disyllabic word and 1 quadrisyllabic phrase entries beginning with jié 孑, most of which can be explained by the meanings in Kroll's Student's Dictionary (see the Appendix for additional details)  There are no word or phrase entries for jué 孓 in Hànyǔ dà cídiǎn, 4.177a.  Some sources say that jué 孓 can mean "short".

To conclude, I would like to introduce LL readers to a major lexicographical undertaking that they likely never heard of, namely, the Harvard-Yenching Institute Chinese-English Dictionary Project.  This was to be a mega-dictionary on historical principles.  So far as I know, only two fascicles were ever published:  39.0.1 (Cambridge, Massachusetts:  Harvard University Press, 1953) and 39.0.2-3 (Cambridge, Massachusetts:  Harvard University Press, 1954) — at least those are the only two fascicles I was able to obtain.  I used to joke that, if they ever really did finish this dictionary on the scale of fasc. 39.0.1 and 39.0.2-3, it would have occupied the whole inner reading room of the Harvard-Yenching Library.

Not to prolong the suspense unduly, 39.0.1 is on zǐ 子 ("child; son; seed").  It has 68 double-columned pages of all known terms and phrases beginning with zǐ 子 ("child; son; seed"), plus a 27 page bibliography.  This, of course, was in the days before computers, and there wasn't even a typewriter or printing press available to Harvard-Yenching Institute that could handle Chinese characters, so all the thousands upon thousands of sinographs were painstakingly written in by hand (pen and ink).

And 39.0.2-3?  It is on none other than jié 孑 and jué 孓!  And, believe you me, the coverage is exhaustive.  19 double-columned pages, of which 16 are devoted to jié 孑 and the remainder to jué 孓, plus a 6 page supplementary bibliography. The treatment in 39.0.2-3 covers graphic variants, phonological variants, and semantic diversities.

Appendix of paleographical and inscriptional evidence

(by Matt Anderson)

Your ideas about 孑孓 are basically the same as the ones I’ve had. I looked into it a little bit more just now, and the situation is a little messier than I expected (and it’s possible that all instances of 孑 aren’t all directly related to each other), but I think there’s good reason to believe that your intuition is at least fundamentally correct.

I can’t find much palaeographical evidence. The only clear example I’ve found is the 孓父乙爵, a mid-Western Zhou vessel so named because it is inscribed with the graphs 孓父乙. So not much to go on there, and not clearly related to any future uses of 孑 and 孓. Note, too, that it’s jué and not jié—otherwise, essentially every use I’ve found of 孓 is combined with 孑 (or is from a dictionary or something).

There are also seals that references books say include 孑 and 孓 like Xihui 0468 (王孓) and Xihui 2999 ([冂+責]孑), but in both of those cases, the graphs in question aren’t really 孑 and 孓 but instead [疒+孓] and [疒+孑], respectively.

I then looked through the early uses included in Hanyu da cidian. Here’s what I found, in rough chronological order (word followed by rough definition followed by source):

1. 孑孑 outstanding, prominent (詩經)
2. 孑遺 leftover, trace (詩經)
3. 孑 a kind of halberd (左傳; this is also supposedly defined in the 方言, but I can only find references saying that it’s in there, not the passage itself in the Fangyan)
4. 孑然 all, whole (國語)
5. 孑 surplus, remainder (方言, the full passage reads “孑、藎，餘也。周鄭之間曰藎，或曰孑。青徐楚之間曰孑。自關而西秦晉之間炊薪不盡曰藎。” So that suggests at least something of a southern connection)
6. 孑/孓 no right/left arm (說文解字)
7. 孑 single, alone (used by Han writers like 張衡 and 孔融)
8. 孑盾 small shield for chariot (from 孑 ‘small’? this otherwise doesn’t seem to be attested early) (釋名)
9. 孑孓 (the earliest appearance I can find of this term is in Guo Pu’s annotation of the Erya—the immediately preceding (and presumably earlier) text says that 孑孑 (jiéjié, not jiéjué) is another word for yuān 蜎 and xuān 蠉 ‘mosquito larvae’)

So what does that leave us? I think the inscriptional evidence mentioned above could easily be unrelated, so I’ll ignore it. Otherwise, 孑 is used in reduplicated form to mean ‘outstanding’ in the Odes, so that sounds like a likely sound borrowing. The Shuo wen definitions can be ignored—Xu Shen needs to define each half of a bisyllabic compound, but there’s no evidence for either of those meanings existing otherwise (at least not until Xu Shen created them). (2) and (5) seem to be the same basic meaning—but the 遺 of 孑遺 already has that meaning on its own—the 孑 prefix in the Odes could be used for sound, or it could be related to the (southern?) term that Yang Xiong mentions in the Fangyan with the same meaning… Or the Fangyan meaning could derive from the Odes usage. (8) could also be related. (4) and (5) also seem possibly related—but the 孑 of 孑然 could easily be being used for sound alone, with the later Han uses deriving from the Guo yu. That leaves (3), about which I have no idea.

And also (9) of course. My guess (and it’s only a guess), based on the above, is that 孑 was used for its sound value and nothing else—from the Shijing through the Han and beyond. When characters were needed to write jiéjué ‘mosquito larvae’ (which certainly seems likely to be a borrowing from a non-Sinitic language), whoever created the form just grabbed 孑 and added a reversed 孑 (孓) to make a phonologically and graphically appropriate written form of mosquito larvae.

But there’s a lot of guesswork going on there.

May 15, 2018 @ 8:59 pm · Filed by Victor Mair under Borrowing, Epigraphy, Language and biology, Philology, Writing systems

Permalink