Hans-Peter left a comment on my previous post asking:

I would be interested how one can learn characters in an efficient way. What do you think about Khatz’ method (Heisig books plus SRS)? Are there better ways? How do/did you learn characters?

My response ended up being too long for a comment, and since it was a topic I planned on covering eventually anyway, I decided to make it today’s post.

Remember the Hanzi and/or Kanji

Remembering the Kanji

James Heisig’s Remembering the Kanji (and its two companion volumes) are mainstays among students of Japanese (John Pasden wrote about his experience with it here, and Mark from Doubting to shuō wrote about using it after having studied Japanese and Chinese for quite a while here). He recently ported the method over to Chinese with Remembering the Hanzi (in both traditional and simplified, with a second volume in the pipeline for both).

The Heisig method is pretty straightforward. Each character is assigned a keyword and a mnemonic story. More complex characters, composed of smaller components, have mnemonic stories that include the keywords for their component parts. Your job is to be able to write the character from memory when given its keyword and mnemonic story. Importantly, you don’t learn the character’s reading at the same time — you save that for after having learning to write all of the characters. In this very interesting piece, Heisig himself gives the reasoning for this:

I told them to look at the Chinese and Korean students at the school. They don’t know any of the pronunciations but they come in with a big advantage and jump ahead of you. Chinese grammar is completely different from Japanese, but Chinese students know the meanings of characters. I want to give myself the same edge as the Chinese have.

James Heisig

I can certainly vouch for this. The characters used in modern Chinese are more or less a superset of those used in modern Japanese, and Chinese students have a huge advantage when starting with Japanese.

Khatzumoto’s method of learning the kanji/hanzi that Hans-Peter asked about is outlined here, and basically mates the Heisig books with SRS to make sure that you review the characters when you’re supposed to. If I were starting to study Chinese or Japanese today, that is absolutely the method I would use to learn the characters — I’m not sure it’s the best possible way to learn the characters (it’s hard to make absolute statements like that for something as complex as human learning) but it’s the best I’ve seen.

How I learned the hanzi

That said, that’s not how I learned to read and write Chinese. I didn’t know anything about Heisig or SRS at the time (2002-2004), and so I put together a method that worked for me from the tools that I had. One important thing to note is that, from the middle of 2003 onwards, I was living in China, and the method I’ll outline below almost certainly relied on being surrounded, constantly, with characters.

My first contact with characters was in 2001, when I came to Changchun for a summer. I didn’t speak a word of Chinese when I arrived, and four months later I more or less still didn’t. I learned a few super obvious characters — 口, 人, 一二三, stuff like that — but otherwise I was totally illiterate. Not fun.

I returned to university for my last year, and took two semesters of Chinese. Though I was planning to go back to China, I still wasn’t a particularly good student, and probably came away with long term knowledge of another 20 or 30 characters. Most of my character learning was through cramming for weekly quizzes, which isn’t exactly a recipe for long-term success.

I graduated from university in 2003, and came to Hangzhou that August. I was planning to stay for a while, and knew that I didn’t want to be illiterate. At first I tried to just learn characters with a textbook and incessant writing and rewriting of characters, but soon realized that at the rate I was learning it would take me forever to get enough characters under my belt to be useful. Soon afterwards, though, I made the best move I could have made and bought the Pleco dictionary for my Palm PDA. At the time Pleco wasn’t nearly as cool as it is now, but it was still a major step forward for my learning. Pleco has two major features that make it killer for learning characters:

1. Awesome handwriting recognition that is very tolerant of mistakes. This is really important when you’re first starting out and proper stroke order has yet to become a reflex.
2. Decent flashcards that integrated with the dictionary, letting you add dictionary entries to your flashcards with a single tap.

I started carrying my Palm everywhere, and looking up characters that I saw on signs, etc., and adding them to the flashcards. One of my favorite pastimes on the slow Y8 bus from the school where I taught to downtown would be seeing if I could look up an unknown character out the window before the bus started moving again (probably the only time that Hangzhou’s terrible traffic helped me out). This didn’t help my writing too much, but it hugely expanded my ability to read.

After that initial period, and following my discovery of Supermemo, I started adding characters to my SRS collection and learning that way. I’ve experimented with the Heisig keyword and mnemonic story approach, but I’ve never stuck with it, though I honestly don’t know why.

Though I’m pretty happy with the results of the way I used to learn the characters, I will admit that my writing is weaker than it should be, given the length of time I’ve studied, though my reading is extremely solid. I’ve always felt that I could patch up my writing pretty quickly given the depth of reading knowledge I have, but I’ve not yet had reason to do so.

Related posts:

1. More thoughts on Heisig
2. Giving Heisig a shot
3. AJATT for Chinese
4. The great debate: simplified v. traditional
5. Repetitio est mater studiorum

Tagged as: chinese, hanzi, james heisig, japanese, kanji, khatzumoto, languages, learning, methods, remembering the hanzi, remembering the kanji, srs