The following comments from Bathrobe in this thread are so interesting I thought they deserved greater prominence:

But in fact there are also cases where the Chinese have borrowed purely Japanese words into Chinese. The mechanism of borrowing is fairly simple: the Japanese write many native Japanese words with Chinese characters; the Chinese feel free to adopt them into Chinese precisely because they are written in characters. Well-known examples are 手続き te-tsuzuki ‘procedures, formalities’, borrowed as 手续 shǒuxù, and 取り消す tori-kesu ‘cancel’, borrowed as 取消 qǔxiāo. In looking at bird names, I’ve also discovered that quite a few Japanese bird names have been borrowed into Chinese, again based on kanji usage — including cases where the Japanese applied existing characters to different birds from the original Chinese, or where they created new characters that didn’t originally exist in Chinese. All of these have been taken into Chinese as though they were Chinese words.
One example that mystified me for a long time was 鶯 yīng. In Chinese this traditionally refers to the oriole or 黃鶯 huáng-yīng. But modern Chinese dictionaries give as a second meaning ‘member of the Sylviidae’ (i.e., the warblers). The extension of the word for ‘oriole’ to the warblers makes a certain kind of sense, but is still mystifying — until you look at the Japanese.
What seems to have happened is that the Japanese took the character 鶯 and applied it to their own cultural equivalent of the oriole, namely the uguisu (scientifically known as the Cettia diphone), a bird celebrated in Japanese poetry for its beautiful song. Thus, the word uguisu came to be written with the character 鴬 in Japanese. The uguisu eventually gave its name to the whole family of Sylviidae, 鴬科. Under Japanese influence, Chinese ornithologists then appear to have applied the character 鶯/莺 to the Sylviidae and thus to the many species of warbler. Since 鶯/莺 is an old Chinese character, this kind of influence from Japanese goes right under the radar of most Chinese.

I had known about words like denwa ‘telephone’ made from Chinese components that were borrowed as wholes into Chinese, but this was completely new to me and is an excellent example of the less obvious influences of the Sinitic writing system on linguistic development.


  1. And bureaucratic words like 引渡 (yǐndù, extradition) and 取缔 (qǔdì, prohibition, crackdown) from 引き渡す (hikiwatas-) and 取り締まる (torisimar-). If my memories serve correctly, they are borrowed by late-Qing government officials from their Japanese collegues.
    Indeed, most such words don’t seem right when you ponder them over, trying to make sense of the word from the morphemes. 取消 (to cancel) and 取缔 (to crack down) basically means 消 (to erase) and 缔 (to close); the prefixing of a semantically rather void 取 (to take) sounds alien. Chinese words with 取 in the first position are usually VO: 取信 is to take trust, i.e. to behave in a way that engenders trust from other people, not a whatever semantic specialization of “to trust”.

  2. komfo,amonan says:

    Wait. When the Japanese use kanji for native Japanese words, why or under what circumstances do they slip the hiragana in there? Do they serve some morphological purpose?

  3. @komfo,amonan
    Yes, they do serve a morphological purpose. Japanese verbs and adjectives use hiragana to show their inflectional morphology. So 行く /iku/ ‘go (plain, non-past)’ contrasts with 行った /itta/ ‘went (plain, past)’ 行きます /ikimasu/ ‘go (polite, non-past)’ etc. This use of hiragana (called ‘okurigana’) also serves to distinguish different words written with the same character: for example, 重ねる /kasaneru/ ‘to pile up’ vs. 重い /omoi/ ‘heavy’ vs. 重み /omomi/ ‘weight, importance’ vs. 重り /omori/ ‘weight, sinker,’ and so on.

  4. And one more syllable when there is a transitive-anticausative alternation. So for 締まる (shimar-, to be shut) ~ 締める (shime-, to shut (a door), the ma part in 締まる is (optionally according to the dictionaries) spelt out, even if not containing the last phoneme of the stem, to highlight the fact that the word exists in alternation with the transitive 締める.

  5. komfo,amonan says:

    @TLO: Oh thanks!

  6. Max Pinton says:

    And then there are the loans that didn’t take, like 地下鉄 (chikatetsu), which means subway in Japanese but only “underground iron” in Chinese. Or least that’s the example I’ve read. Maybe by now it means subway in Chinese too.

  7. One modern example is お宅 (otaku) turning into 宅男/宅女, though from experience the meaning of 宅男/宅女 has deviated (quite a bit?) from the common Japanese meaning.
    For example compare the common conception of a Japanese otaku to Jay Chou’s 阳光宅男 song:

  8. @p
    What is the meaning of 宅男/宅女? I have little functional knowledge of Chinese, so the video doesn’t help a lot.

  9. Basically hikikomori, and more generally, people who spend a lot of time at home on Internet.
    地下鉄 means subway to a lot of Chinese, though more Japanese and therefore cooler.

  10. 地下鉄 means subway to a lot of Chinese, though more Japanese and therefore cooler.
    Funnily enough, while visiting my local Asian grocery store I’ve noticed that many packages from China (and some from Thailand) seem to have Japanese written on them for no practical reason at all. Seen on one bottle of green tea from Thailand, prominently displayed: おいしい (oishii delicious) 健康 (kenkou health) 緑茶 (ryokucha green tea). And that was the only Japanese on the package (the rest was in Thai and, thankfully, English).
    Japanese definitely seems to have a certain coolness associated with it in East and Southeast Asia, despite all the terrible recent history.

  11. @TLO
    宅男 zháinán/nǚ means something very similar to the Japanese term otaku, that is, someone who will frequently watch anime, read manga, obsessively collect items or paraphernalia related to that subculture, etc. The tendency is for people who are called “otaku,” or self-associate as an otaku, to be socially awkward or inept. (For more information, Neojapanisme has some of the original essays on the topic translated:
    However, on the Chinese side, the term 宅男/女 has been adopted to be much more broader. A girl could say “Oh, I’m going to have a 宅女 night” – implying that she is simply going to stay home instead of socialize. A man can be a 宅男 if he is shy around girls or does not actively seek out social engagements. Neither of the terms or usages imply that the person has any association with the sub-culture implied by the Japanese usage of the term. Of course, you will occasionally hear the term used in China to mean that, but I’ve only encountered it rarely.
    The Jay Chou video sort of shows what a (at least American) stereotype of a geek might be – contrasting with the potentially obsessive Japanese otaku subculture.
    Hopefully that makes sense (and wasn’t too long winded)!

  12. My favorite example of this is 可愛 kě’ài from 可愛い kawaii, especially since the choice of kanji is based on the mistaken idea that kawaii is derived from Chinese elements in the first place.

  13. 可爱 is attested in premodern Chinese. Every schoolboy knows “风移影动,珊珊可爱” from 项脊轩志 from Gui Youguang (1506-1571).

  14. minus273 has cool (if sporadic) blogs. But they would be even cooler if he consolidated them!

  15. Actually, 自動車 jidōsha wasn’t borrowed into Chinese. The Chinese word for ‘automobile’ or ‘motor car’ is 汽車 qìchē, which normally means ‘train’ in Japanese.

  16. Thanks Bathrobe. I’ll definitely try to update more when I pulled myself out of my mud-pool of syntax devoirs to do in LaTeX.

  17. Actually, 自動車 jidōsha wasn’t borrowed into Chinese.
    Thanks, I deleted it from the post so as not to mislead people. That’s what I get for trusting Wikipedia.

  18. Here’s a list (compiled by Victor Mair) of “East Asian Round-Trip Words” (starting on page 7 of the pdf):

  19. (typically Archaic Chinese > Japanese > Mandarin Chinese, with the 2nd Chinese definition quite different from the first)

  20. I love this kind of stuff, especially the taxonomic adaptations. I’ll think of an oriole now every time I hear the cry of the uguisu in Japan (especially when it’s the recording played over the PA system in busy Kyoto Station).
    I understand that a majority of the Japanese kanji for fish names are kokuji ‘national characters’ (i.e., made in Japan), which makes a lot of sense when you compare the different kinds of fish that turn up in Japanese vs. Chinese diets (and waters). I wonder how many of those have now been borrowed into Chinese and what role Taiwanese might have played in contemporary J > C borrowings.

  21. I’ve had a quick look at the very start of East Asian Round Trip Words. The list is a bit misleading as it stands without interpretation.
    For a start, the process by which calques were formed raises questions about whether these were really ‘borrowed’ from Chinese. 文法 might have meant ‘civil rules’ in ancient Chinese texts, but when the Japanese formed the calque, it seems that they deliberately ignored the original meaning, choosing to interpret 文 in the meaning of ‘literature’ or ‘literary text’. Can it really be said that 文法 made the trip to Japanese, changed meaning, and then came back to Chinese? Wouldn’t it be possible to say that the Japanese created the word 文法 according to Chinese rules of word-formation, and then exported the resulting form to Chinese? Of course, familiarity with the old Chinese term 文法 may have acted as a ‘model’ (unconscious or otherwise) for creating the form that carries the new meaning.
    I also noticed that 文明 means ‘patterned brightness’. This looks like one of those cases where ancient Chinese texts used the same character in several different meanings (the only, rather obscure, example that springs to mind is 菟, where 菟 could mean either ‘rabbit’ or ‘a kind of plant’ — whereas in the modern language, only 兔 is used for ‘rabbit’). In this case, 文 looks like it was used for both 文 and 紋 in Classical Chinese, 紋 having the meaning ‘pattern’. I can’t check this at the moment, being short of reference materials.
    Neither of these points detracts from the interest and usefulness of the list, but a lot of the mystery that surrounds these words when presented baldly without comment is capable of elucidation.

  22. I’m not sure what Joel means by “contemporary” J-C borrowings. My gut feeling is that the era of great borrowing was the end of the Qing/start of the Republic, when many Chinese went to Japan to study. I suspect that they were instrumental in ensconcing the new Japanese vocabulary in Chinese. Prior to that there were a number of different attempts to render Western concepts in Chinese. The Chinese are particularly fond of eulogising Yan Fu as the great translator of the late Qing era, and yet Yan Fu’s translations of Western concepts have virtually all been discarded in favour of these ’round-trip words’ from Japanese.
    I’ve noticed that Japanese usages do seem to have been more widely retained in Taiwan Mandarin (and possibly Hong Kong, etc.) For example, the Japanese word for an astigmatism, 乱視 ranshi appears to have survived in Taiwan as 亂視 / 乱视 luànshì whereas the Mainland has standardised on 散光 sǎnguāng. Or the preference for 連絡 / 联络 liánluò in Taiwan and (I think) HK, where 聯繫 /联系 liánxì is preferred on the Mainland (esp in the north). Japanese is, of course, 連絡 renraku.

  23. Sorry to harp, but the ‘Round-Trip Word’ 鉛筆 enpitsu is also misleading as it is presented.
    The Chinese meaning is given as ‘lead writing instrument’, whatever that might have been. Did they use lead writing implements in thosee days?
    The gloss for the Japanese calque, ‘[graphite] pencil’, is quite inaccurate. 鉛 does not mean ‘graphite’; 筆 does not mean ‘pencil’. The meaning of the Japanese is exactly the same as the Chinese: 鉛 means ‘lead’, and is based on English ‘lead pencil’ (the normal although inaccurate term for it); 筆 means ‘brush’ or ‘writing instrument’. So the Japanese took ‘lead writing-instrument’ and applied it to the lead pencil. The big transformations in meaning implied by the gloss simply didn’t take place.

  24. FotD: There’s only one major deposit of pencil-quality graphite in the world, and it’s in England. No others have ever been found. (Pencils nowadays are made of powdered graphite held together with glue.)

  25. My mistake. Mair’s list is internally consistent. The gloss he gives to calques is not a literal gloss; it’s a gloss of what is being calqued. 鉛筆 enpitsu does literally mean ‘lead writing-instrument’, but it has been adopted as a calque for ‘[graphite] pencil’. However, using the less-correct original term ‘lead pencil’ would make the connection clearer.

  26. By ‘contemporary’ I meant the period when Japanese menu items, particularly fish names, began appearing in Chinese restaurants, kitchens, and store shelves after China began opening up in the 1980s, as people from Japan, Taiwan, and HK began doing more and more business on the mainland.

  27. Not sure about fish names so you’d need someone else to fill you in on that.
    Tatami (榻榻米) has, I think, entered via Taiwan. ‘OL’ has also tentatively entered the parlance of young women in Beijing, but I think it was via Hong Kong. There is probably a lot more that I haven’t noticed.

  28. wad.
    For a while, the Thoreau family business got theirs from one of the next best deposits, in Sturbridge, Mass. (Metewemesick).

  29. Fish names in Taiwan can be a nightmare for the translator. They are often different from both the Japanese and mainland Chinese names, and sometimes there are “local” or common names (俗名) in use as well. Those might be unofficial transcriptions from Taiwanese, names used in the marketplace, etc. A handy database is:
    From there, you can look up the Chinese to get the Latin, and with the Latin you can figure out the English.

  30. John Emerson says:

    Birds and fish have local names everywhere. Here in MN there are two unrelated fish called “dogfish”, one a burbot (freshwater cod) and the other a bowfin. In all there are over a hundred species called dogfish, freshwater and salt. The name is the official name for one or more species of shark.
    Cisco, chub, lake herring, tullabee, and whitefish are the local names for several different species of commercial fish in the Great Lakes and Midwest, salmonids in the Coregonus artedi family. To my knowledge there are no matches between the several species names and the several popular names; it’s an all-to-all mapping.

Speak Your Mind