A while back Albert of Laowai Chinese visited Shanghai. We met up for lunch and had a good chat about our experiences in China learning Chinese. He asked me an interesting question: what did I think was the biggest problem with the field of Chinese language instruction?
I told him that in general, I felt that there was way too much teaching adult foreign learners as if they were Chinese children, and I felt that more (non-Chinese) learner perspectives were needed to improve the situation. (This is one of ChinesePod‘s major strengths.)
He was looking for more specific answers, though. When pressed, I gave him these two areas:
Tones should be taught systematically, long-term. Way too many programs cover the tones in the first few weeks, followed by a few tone change rules, and then basically leave the students to sort the rest out. It’s not enough, and it’s irresponsible. Most students are going to need a good 1-2 years to really get a handle on the tones, so why aren’t educational institutions doing more to guide students through those frustrating times?
As I’ve said before, tones were the single most difficult part of learning Chinese for me, and I know it’s true for many other students as well. More needs to be done. We make this a major focus at AllSet Learning, but most schools really drop the ball on this one.
Mandarin Chinese needs a public, large-scale corpus of spoken Mandarin. There are corpora for Mandarin, but the ones that are public are not spoken Mandarin, and the corpora of spoken Mandarin are kept private and jealously guarded.
Why does Mandarin need a public, large-scale corpus of spoken Chinese? Because without it, we’re all just taking stabs in the dark as to what “high-frequency” spoken vocabulary is. Yes it is possible to objectively determine what language is high-frequency, but this requires (1) collecting lots of naturally-occurring speech samples in audio form, (2) transcribing it all. Then a proper corpus can be assembled, from which accurate, objective word counts and word frequencies can be derived.
Once that’s done, we could finally have more of a clue as to what the “high-frequency” spoken vocabulary really is. This method isn’t perfect, but it’s a big step forward from relying on native speaker intuition. And no, the new data obtained are not going to match the HSK word list you’ve got, or the Jun Da list either.
It would also be great to see a proper large-scale corpus of spoken Mandarin, balanced for regional variation. That would turn up all sorts of interesting facts, like proportion of 哪儿 to 哪里 across all regions represented, and virtually any other speech variation you can think of. (Personally, I suspect that a lot of the Beijing-hua taught in many textbooks could be reconsidered on the grounds that it simply doesn’t represent the Mandarin spoken across mainland China.)
What do you think are the biggest problems with Chinese language instruction today?