Discounts on non-Sinitic Languages
In a post written by Syz in June, the topic of language discounts was discussed. By all means, if you haven’t read it, go take a look at the post and the 30+ comments it brought about. I want to review this idea for a moment, but this time on non-Sinitic languages. Specifically, I want to look at Korean.
As I mentioned in my last post, I’m giving myself a crash course in Korean. I’ll be in Seoul in a couple months and as nice as my friends there are, I don’t want to have to rely on them for conversations on topics like hotels, taxis, restaurants. In fact as added incentive, my friend has given me a list of things I must master by the time I get there. This includes the above few tourism-related areas, but then also I’m expected to learn at least one pop song for karaoke (Kr: 노래방 noraebang, lit. “song room”). I’ve only just begun, and I’m doing what I can to make sure my pronunciation is passable from the start. But I can’t help thinking I’m about to hit a stage in my vocabulary acquisition where it just explodes into awesome. It has a lot to do with discussions had over at Annals of Wu on what I call the “sound matrix” theory.
The idea, which I should add is invalid, is as follows:
Language X is essentially Language Y with a set matrix of sound changes applied, and, if the learner knows Y and then simply learns those changes, they will be conversant in X.
With Wu this means you learn a few basic rules such as “words ending in -ang in Mandarin will be -ã in Wu”. But again this doesn’t actually work out. For starters, Wu has more phonemes available than Mandarin. Where we have [p] and [pʰ] in Mandarin we find [p] [pʰ] and [b̥] in Wu. So obviously we can’t say with certainty which sound will end up as which. Of course, it does work sometimes. I was one in a Hebrew class, having studied Arabic extensively, and when needing the word for “dog” in Hebrew (כלב kelev) I was able to guess with some accuracy based on the Arabic (كلب kalb). I couldn’t guess how often it does actually work out, but I’d say less than half the time. That said, it is of some use.
It turns out that in Korean, which according to the Lonely Planet guide has a vocabulary that is 70% Sinitic (as compared to the anecdotally 60% of English words that are derived from French), that <50% chance of getting it right may be significantly higher when going from Mandarin to Korean.
Just for example, the 방 bang in 노래방 noraebang is cognate with Mandarin 房 fáng. Were we to start codifying the sound matrix, we could perhaps say that Mandarin syllable-initial /f/ becomes /b/ in Korean cognate words. Indeed in the Hippocrene Practical Korean-English / English-Korean Dictionary that I was recently gifted, a great many of the entries have the equivalent Sinitic characters (漢字/한자) for entries borrowed from Sinitic languages. Opening to a random page we get this as an entry:
시인 詩人 a poet (p.306)
That’s ʃi-in in Korean and shīrén in Mandarin. Knowing a little bit of Wu, or at least how the matrix would go in the few cases in which it works, is of some value as well.
As I mentioned, this may explode in a couple weeks. I’ve been making a mental list of the common changes, though to call them changes isn’t necessarily accurate, and I’ll soon compile it all into one list. Some notable ones is the shift of /f/ to /b/, /ʃ/ to /s/, initial /w/ to /m/ which we find in Cantonese and Wu in most cases as well, both of which are holdovers from Middle Chinese.
It is of course no big surprise that Korean has so many of these such holdovers. It’s been used by scholars such as Kalgren and Pulleyblank in their reconstructions of Middle Chinese pronunciation. The notable part for me at least is that there seems to be some real value in being aware of such similarities.
I’ve often thought there would be use in creating simple guidelines if not entire curricula for language classes that would teach Hebrew for Arabic students or Cantonese for Mandarin students, or in this case Korean for Mandarin students. It’s a niche market to be sure, and for the most part the kind of people who would read this blog are likely good enough at making those connections for themselves that perhaps it’s completely unnecessary. But I for one certainly wouldn’t mind having a guide in this process, someone who’s done it before, if for no other reason than efficiency given the short time I have given myself.
I’m interested if anyone has had this experience between Mandarin and another language or really any two other languages for which the discount may have been unexpected. I had assumed what little work with Manchu I’ve done under the guidance of Paweł and Randy would be of greater value in learning Korean than anything I’d done with Mandarin. And who knows, perhaps when it comes to grammar that will prove to be the case.
I’m fluent in Japanese and currently learning Mandarin. I often find myself being able to guess Mandarin hanzi readings based on Japanese, and sometimes can parlay this into entire words. Of course this only works with the (numerous) Chinese loanwords that exist in Japanese; there’s pretty much no connection between “native” Japanese words and Mandarin (房 is used in words like 厨房 chūbō, but the native word for “room” is heya or ma).
Japanese has such a “coarse” set of phonemes that any conversion matrix between Mandarin and Japanese would have to be on the level of syllable. For instance, some hanzi read yuan in Mandarin have the pronunciation en in Japanese (园圆缘苑), xiao in Mandarin → shō in Japanese (小笑肖消宵), etc.
Aaron:
One of the nice things about words borrowed from Mandarin or some recent ancestor is the small range of sounds used by Mandarin leading to potential uniformity in the target language. I don’t know much about Japanese but I would guess the fact that various readings exist for beach character would make it much harder to predict readings were one learning Japanese after mandarin instead of the other way as you’re now doing.
I wonder if this would actually be more helpful in later language learning. English tends to retain the native English words for “core” vocabulary and use French borrowings for more “educated” vocabulary. I wouldn’t be surprised to learn that this is also the case for Korean and Chinese.
Either way, yes, I think language teaching materials would do well to make the parallels explicit. I don’t know how many Chinese people there are learning Korean, but at least where I am, there are a lot of Korean speakers learning Chinese.
Maybe, I’m understanding this incorrectly, but apparently Korean is a language isolate according to Wikipedia. However, it might also be related to the Altaic language tree, which includes Japanese.
Thus, I find it strange that you could relate Korean to Chinese. Is this only because of borrowings? Or something similar that I’m missing here?
But in terms of a guide that’ll help Mandarin students learn Korean, I’ll would definitely like that, as when I solidify my Mandarin, I would like to move on to other sino languages, perhaps Japanese or Korean.
Oh silly me, I seemed to have skimmed on the title. Non-sinitic. I get that now. Still interesting comparisons though!
The reverse of Aaron’s experience is also totally true, especially if you’re just going for meaning. I can follow relatively complex topics in a Japanese newspaper with pretty elementary Japanese because of the kanji, though knowing the pronunciation seems only to be a memory aid for remembering the pronunciation — it’s seldom seems close enough to really guess.
As for other languages, the work I’ve put in in Japanese has made picking up the basic Korean and Mongolian that I know a lot easier. I think it’s that I’ve already done the mental exercise needed to deal with verbs following their objects, which was pretty mystifying at first with Japanese.
Prior knowledge of Cantonese is a great help in memorizing (Sino-)Korean vocabulary because both Cantonese and Korean have preserved the old -p/-t/-k and -m finals (although -t historically changed to -l in Korean). Some compound words are pronounced almost exactly alike in Cantonese and Korean, e.g. 雜誌 “magazine” (jaapji/japji) or 合作 “collaboration” (hapjok/hapjak).
It seems to me that in the few cases where Cantonese and Korean finals diverge, it is generally Korean that has retained the historically “correct” pronunciation. For example, 核 is hat6 in Cantonese but 핵 (haek) in Korean; according to Pulleyblank, 核 originally ended in -k.
In my own experience, a background in Mandarin is occasionally useful in learning Korean, but there’s nothing systematic about it. (I understand it gets much more useful and systematic for more formal language.)
A background in Japanese is quite useful for Korean grammar, to the point where I was getting really impatient “relearning it” and just wanted a guide to Korean for the Japanese speaker.
I figure you can do the same for Vietnamese, too — there’s a lot of terms with Chinese cognates. Take “Socialist Republic of Vietnam,” which is Cộng hòa xã hội chủ nghĩa Việt Nam, corresponding to Gonghe Shehuizhuyi Yuenan in Mandarin Chinese.
Also, not quite related, but it’s possible to pick up tones quickly in a Sinitic language given prior knowledge in another (although the ease of this process depends on the exact pairing, of course). For example, a Cantonese speaker can map 6 of the 9 tones (those that are found in syllables that don’t end in -p, -t, and -k) to all 4 tones of Mandarin Chinese pretty accurately, because these tones all descended from the same set of tonal ancestors. Of course, the monolingual Standard Mandarin Chinese native speaker, having less distinctions in tones, will find the mapping to Cantonese trickier.