In a comment on the previous entry, Joel of Far Outliers mentioned a post he’d done on acronyms in Chinese and other Asian languages that were influenced by Chinese:
In Chinese, for instance, acronyms are composed of the initial syllabic characters of (usually) two-syllable words. So, Peking (= Beijing) University, or Beijing Daxue [lit. ‘NorthCapital BigSchool’] becomes Beida [lit. ‘NorthBig’]. In Korean, Korea University, or Koryo Taehak [lit. ‘HighBeautiful BigSchool’] becomes Kodae [lit. ‘HighBig’]. In Japanese, it’s a bit more complicated. Chinese characters can be pronounced not just in their Chinese loan forms, but as native Japanese words that mean (more or less) the same thing… So the acronym for Hiroshima University, or Hiroshima Daigaku [lit. ‘WideIsland BigSchool’] becomes HiroDai [lit. ‘WideBig’]. The name Hiroshima is native Japanese (the Sino-Japanese pronunciation would be Koutou = Ch. Guangdao), but Daigaku is borrowed [= Ch. Daxue].
He goes on to give examples from Indonesian and Vietnamese (which uses initial letters, like English, rather than combining initial syllables). I would direct the interested reader to my own entry on the phrase gung ho, which has its origin in precisely this form of abbreviation.
A side issue: the remark “Daigaku is borrowed [= Ch. Daxue]” is presumably true, but I wish there were a site where one could easily learn which common words were borrowed one way and which the other—a great many words relating to modern phenomena were coined in Japan and then borrowed into Chinese via the characters, for example Japanese denwa ‘telephone,’ borrowed into Chinese as dianhua. (There’s an interesting discussion here, which unfortunately gets sidetracked by a pointless argument over whether the formation of the Japanese word is important.)
My understanding is that one person, Fukuzawa Yukichi, the famous Meiji-era Westernizing journalist and founder of Keio University whose image grace the Japanese 10,000 note, is reponsible for hundreds of Sinitic neologisms that were easily transferred into Chinese and Korean. I suspect he might have coined Jp. Daigaku, which engendered Ch. Daxue (and indirectly perhaps even Hawaiian Kula Nui ‘School Big’).
One source I haven’t read that might be promising is a recent book called Translating the West: Language and Political Reason in Nineteenth-Century Japan, by Douglas R. Howland (U. Hawai`i Press, 2001).
I edited “Daigaku is borrowed [= Ch. Daxue]” to “Daigaku is Sino-Japanese [= Ch. Daxue]” in order not to imply anything about directionality. In a few cases, it’s clear that Ch. and Jp. neologized separately: ‘train’ is Ch. huoche ‘fire-cart’ but Jp. kisha ‘steam-cart’, while Ch. qiche ‘steam-cart’ means ‘automobile’. I think ‘bicycle’ is also different in each language.
My favorite Japanese loan word in Mandarin is the word for karaoke. It is purely phonetic, but because Chinese doesn’t have some of the sounds, they borrow from English as well! It is the characters for “ka” and “la” followed by the English letters “OK”: ka-la-OK!
Note: Hoklo/Hokkien/Taiwanese/Southern Min (whatever you want to call it) has lots of Japanese loan words. 50 years of colonial rule left its mark on the language in a way that it didnt’ in Mandarin. Words for truck, radio, chocolate, motorcycle, neck tie, etc. are all very close to the Japanese (and these words are all loan words from English in Japanese…).
I wish there were a site where one could easily learn which common words were borrowed one way and which the other—a great many words relating to modern phenomena were coined in Japan and then borrowed into Chinese via the characters
I am not aware of a site, but Lydia Liu’s Translingual Practice: Literature, National Culture and Translated Modernity China, 1900-1937 has lists of all the ones she could find
My favorite Japanese loan word in Mandarin is the word for karaoke. It is purely phonetic, but because Chinese doesn’t have some of the sounds, they borrow from English as well! It is the characters for “ka” and “la” followed by the English letters “OK”: ka-la-OK!
Mandarin says Ka la OK too.
When I was in Taiwan I noted a number of ways that the roman alphabet had been integrated into Chinese as quasi-Chinese characters. The two I remember are “F”, abbreviating the English “floor”, being used by my mailman to route my mail: “5 F” is read “wu lou”. Reading th 5 as a Chinese word is normal, but “F = floor = lou” is an innovation.
The other was the use of “W = week = xingqi” so that “W1 = xingqi yi = Monday”.
My understanding is that this was done more for fun than as a timesaver. It really complicates the language rather than simplifying it. The post office was coding letters written in English, but the deliveryman was monolingual in Chinese and there were almost no Westerners in my neighborhood. Most doctors and nurses probably do read English.
The preservation of the complicated Chinese writing system is usually interpreted in terms of class (high literacy marking excellence and achievement), but the playful and fussy aspects shouldn’t be discounted. Fussing around with petty detail is part of the fun.
Hong Kong and Taiwan write “sandwich” differently: “san-ming-zhi” vs. “san-wen-zhi” (Mandarin values). The Nationalist ideology is “San-min-chu-yi” (San Min -ism)and I once asked how someone could make a political system out of sandwiches. Joke not appreciated. The dissidents when I was there called the Nationalist system “toothbrushism” (ya-schwa-chu-yi) though I wasn’t able to get any details.
I got off at the “Qingtan Guoxiao” (Green Pool Nation Small) busstop for a year before I realized it meant “National Primary School”: Guoli Xiaoxue. Incidentally, the word “zhan” in busstop is thought to be derived from the Mongol “yam”. Traditional Chinese before the Mongol civilizing influence had only the most primitive concepts of commuting and would just try to flag busses wherever they saw them.
Slightly off-topic: I’ve heard “OK! Bye-bye” at the end of long conversations conducted entirely in Chinese. It really does seem like a very Chinese kind of thing to say at the end of a conversation — two binomes or a four-character cheng-yu. I think that “bye-bye” might also have been assimilated to the Chinese “bai-bai”, which means to make devotions or pray, and so would be more respectful and less casual than the English.
Correction: “ya-shua-zhu-yi”.
Note: I saw “W1” for “Monday” on a hospital bulletin board, which explains my reference to doctors and nurses.
English does use syllabic acronyms in some cases, e.g. “fro-yo” for “frozen yogurt”, “milspec” for “military specification”. Most of the examples I can think of are either informal or military–the military is a heavy user of acronyms of all sorts.
Not meaning to be pedantic, but for japanese (and I assume Chinese) it isn’t syllabic based, it is character based. When those characters bump up against each other, adjustments have to be made. So Tokyo U is Todai, Hokkaido U is Hokudai and Tohoku U is Tohokudai, even though Tohoku might be considered more prestigious (but Hokudai is more ‘famous’)
That it is character based (which means that meanings are retained) can be seen in the names for expressways and railway lines. The famous Tomei expressway (no relation) is To(kyo)-Na(goya), and Na is read with a kun yomi as ‘mei’ describing the two endpoints of the line.
Joe,
Only in Japanese does character-based not equal syllable-based. And that’s for two reasons: Japanese had to add final vowels in order to pronounce the syllable-final consonants (obstruents) that Chinese used to have; and Japanese assign native pronunciations to Chinese characters. Koreans (and obviously Chinese) don’t.
I would recommend several books on the topic: John DeFrancis’s _The Chinese Language: Fact and Fantasy_ (which argues that Chinese characters are just one huge syllabary, with 800 symbols), and his _Visible Speech: The Diverse Oneness of Writing Systems_ (which argues that *all* writing systems are basically sound-based); and J. Marshall Unger’s _Ideogram: Chinese Characters and the Myth of Disembodied Meaning_. All are published by U. Hawai`i Press.
In Japanese ‘train’ is usually ‘ressha’ (line cart) or ‘densha’ (electric cart), ‘bicycle’ is ‘jitensha’ (self step cart) and ‘car’ is ‘jidousha’ (self move cart) or ‘kuruma’.
In Mandarin ‘dianche’ (jp densha) = tram, ‘bicycle’ is ‘jiaotache’ (foot step cart), ‘danche’ (single cart) or ‘zixingche’ (self travel cart). ‘jiaotache’ is used in Taiwan while the other words are used in China.
Simon,
Thanks. I was comparing the original neologisms for ‘train’ (or ‘locomotive’) in Ch. and Jp. It’s true you don’t see or hear kisha ‘(steam) train’ much in Japan anymore. Densha means ‘tram, streetcar’ in Jp. too (but you hardly see them on the streets anymore–except in Hiroshima, where the delta land makes a subway system difficult). Jp. jitensha ‘bicycle’ is actually ‘self-rotate-cart’, not ‘self-step-cart’.
Zizka,
It sounds like Taiwanese are assigning either native or foreign sound values (or both) to borrowed symbols, just the way Japanese have done for centuries. Another subtle bit of Japanese influence there?
Your note about ‘sandwich’ seems to indicate that the Ch. equivalent was coined when Mandarin held sway, unlike ‘curry’ which is jiali in Mandarin, but a much more likely gali in Cantonese. You can tell that Cantonese coined many of the character-strings used to render early foreign place names, like Mandarin jianada ‘Canada’, jiana ‘Ghana’, and jialebi ‘Caribbean’, all of which start with ga- in Cantonese.
I’ve been wondering about Jianada for years. I guessed it was a dialect thing but I don’t know anything about Cantonese.
DeFrancis stakes out and advanced position and makes a case for it. I ended up non-committal about his thesis — he’s pretty polemical. But then, I’m one of those who likes the fun/fussy aspects of Chinese script.
Here’s a Chinese dialectal place name I’ve been wondering about. When or where did the Peking pronunciation come from (not pee-king, but probably something like bei-king in pinyin)? In Cantonese it sounds like buck-king (archaic Cantonese is closer to Sino-Jp. hokkyo), and in putonghua it’s obviously Beijing (not with the French /j/ that all the newscasters use). So maybe Peking comes from one of the dialects around the southern capital Nanking (now Nanjing) or Shanghai area.
Pospelov just refers it to “a southern Chinese dialect”; I too am curious about the exact origin, but maybe it’s irrecoverable.
Somebody at the U. of Hawai`i (where DeFrancis taught) a few years back did a series of experiments on how Japanese readers process Chinese characters in running text. A series of texts were altered so that the characters in one control set had systematic mistakes in the phonetic portions, and those in the other control set had systematic mistakes in the semantic portions of those same characters. Tracking of eye movements and visual “stumbles” showed that readers had a lot more trouble when the phonetic clues were wrong than when the semantic clues were wrong.
I suspect that’s one thing that makes proofreading pretty tough in Japanese or Chinese.
BTW, the book by Unger entitled Ideogram that I mentioned above has a whole chapter on similarities between Japanese orthographic conventions and Gregg shorthand–both of them relying on radically underspecified phonetic clues and ambiguous abbreviation conventions that cannot easily be interpreted without a lot of contextual clues.
Hi Joel,
I’ve read the two DeFrancis books, and I have to disagree with some of the points (I haven’t read the Unger book, but I’ll discuss that a bit later)
DeFrancis’ first book was a welcome remedy for a lot of the garbage that circulates about chinese characters, and so, his polemic tone can be excused. But when he tries to extend the argument to all languages, there are a number of problems. I will try and note some of them here.
First, if writing systems are so strongly rooted in speech, we would expect illiteracy to be much less of a problem than it actually is. In the strongest version of DF’s claims, we would expect that only those with some sort of neural deficit in speech would be illiterate, but this is definitely not the case. (I would also note that it would be a lot easier for foreigners to learn as well, if they have a sound grounding in the spoken language, but this sadly is also not the case)
Second, if writing systems are encoded speech, why is it that tones are often not encoded. In Thai and Vietnamese, one can know the tones if one remembers a small set of rules, but this is not the case in Chinese. Also, in African tonal languages have been able to get by quite nicely without tone marking. One could argue that tone is a separable component from the rest of phonetics, but research suggests that this isn’t really viable.
Third, DeFrancis tries to argue that sign languages are somehow related to speech. I understand that he wants to slay the idea that characters are simply pictures, he goes to the other extreme by suggesting that sign languages arise out of speech, which then makes sign languages the ‘mirror’ of speech, recalling the kind of nonsense that sign languages are not authentic natural languages.
Fourth, he bases his arguments on modern Chinese, and dismisses the evidence from very early Chinese, most notably Shang Oracle inscriptions. The strong version of DeFrancis’ thesis would make no allowance for the change from a meaning to a phonetic based script, though research strongly suggests an primary iconic component to such writing.
Fifth, even if there is a phonetic basis to the writing system, that does not answer how the writing system is processed in reading.
DeFrancis first book spends a lot of time discussing the Creel-Boodberg debates, taking Creel’s side. This is fine, because it is important to understand the debate between whether Chinese characters are primarily phonetic (Creel) or semantic (Boodberg) in nature. But the debate (from 1936!) is reductive and fails the same way that debates over phonics versus whole-word approaches fail. It is relatively clear (at least to me) that we use a combination and it is not one or the other. The same is true, I would argue, for Chinese characters, though I have no special insight into the mind of a native speaker of chinese. In _Visible Speech_ Defrancis extends the argument to _all_ languages. Yet you suggest that Japanese is an exception and if there is one exception, then the strongest version of DeFrancis’ argument can’t stand. I would recommend Taylor and Taylor”s _Writing and literacy in Chinese, Korean and Japanese_ where he notes that DF’s argument “may have been necssary to counter the opposite extreme view (of chracters as only meanings), but… But logographs and phetic scripts differ in the mannter and efficacy of representing the meanings and sounds of morphemes”
Unger’s book, I haven’t read, however, I think that the Ideogram book is not an academic book, as is evidenced by the title, blurb and cover. (all of which are here This is not to disparage Unger as a scholar, but when you write a book to convince the majority of people of a position, you end up leaving out a lot of the nuance (cf Pinker, esp. _Blank Slate_)
There is a lot on interesting research about visual stumbles,eye tracking, PET activation scans, but they do not have the granularity to give us the exact answers yet. It is easy to sit and pick at someone else’s construct, so I’ll try and offer my own theory herel. My feeling is that language ability (which includes speaking, reading, listening, writing, the whole ball of wax) is like a word processing program like M$ Word. You can sit down and navigate someone else’s setting and get by, but if you really need to get work done, you need to have it set up the way you want it. I imagine that the language ability is not modular and not unitary, but each person has variable strengths and weaknesses and plays off of them in different ways. There is the neurological phenomenon of people who can understand gestures, both spoken and visual, yet have no ability to understand the logical meaning, while there are others who can only understand the logical meaning and are unable to pick up on irony, sarcasm, etc. Oliver Sacks discusses this in _The man who mistook his wife for a hat_ These people are at the extreme ends of the spectrum and I imagine that the population is distrubuted across this spectrum. In Chris McManus’ book _Right Hand, Left Hand_ he also sketches a similar scenario in regard to handedness, noting that even in cultures where right handedness is culturally dictated, there is an irreducible percentage of left handers. I’d like to think that there are readers who are using phonetic clues, and readers using semantic clues, and this accounts for the variable performance.
Sorry for the long post. Look forward to your comments
Please don’t judge the book by its cover. It does have an academic message (see especially chapters 7 through 9), and I certainly didn’t write it with the thought that it would sway the thoughts of millions (although I’m sure my publisher will be delighted happy if it sells well).
Thanks for the information; I’ll try to have a look at it. (You must not have been thrilled when you saw the cover!)
No, a little levity is fine.