Aug 042008
 

EDIT: I put the links in at the bottom of the page, in a more coherent order.  I also added a couple I found recently.

This is something I’ve been wanting to talk about for a while, but I never really got around to doing it for various reasons. But now I want to try to explain my approach to language learning. It’s nothing that I discovered on my own. I just found some people with some ideas that seem to work pretty well for me, that are backed up by some pretty solid science.

I have chosen a few weapons for my language-learning arsenal. The first is the used of a good Spaced Repetition System to commit grammatically correct sentences in the target language to memory. The second is the use of immersion when I can. The third  is the practical application of Stephen Krashen’s comprehensible input theory which emphasizes input over output. Basically, you want to overwhelm your brain with good input in the target language before forcing output. (Output being you opening your mouth and trying to come up with something coherent.)

Fortunately for me, some people have already come across these ideas and fleshed them out. Khatzumoto at the AJATT blog has written extensively on his experiences, the guys at antimoon.com have also achieved some excellent results, as has this poster on kuro5hin.

Dunk My Brain in Some Japanese!

Immersion is pretty straightforward. If you immerse your brain in a surrounding composed of your target language, it will pick up on things much faster than if you don’t. It’s a pretty obvious thing. The more of X language you listen to, the better you’re going to get. That’s how you learned your first language, anyway. Your parents did NOT sit you down with a stack of textbooks and start giving you exams on grammar. You just picked it up. That’s what your brain does.

I’ve been slowly trying to build a more immersive environment around me, but it’s hard to do. I try to watch as much un-subtitled Japanese TV as I can stand, I read books, manga, etc. It helps a lot, but I need to work a little harder at weeding out extraneous English.  But English is my money-maker. That can’t be helped, because I love it, too.

I had my own good example of this when I was in Japan. My Japanese would get better whenever I stayed at a hotel that DIDN’T have English-language programming, and it got worse when it did. (Because I would watch it.)

Without memory, all of this immersion effort is really inefficient. The basic technique I use is called spaced repetition, and by using that method, you can slowly push data deep into your long-term memory by pushing your “forgetting curve” farther and farther back. There’s also a good writeup of how the author of Supermemo gets his SRS on in a quest to fill his brain in this Wired article.

I use an SRS program  to cram as much Japanese into my noggin as I can. My SRS of choice is Anki, which is an excellent SRS program to use for learning Japanese. It works well on PC, Mac, Linux, and over the web. It dovetails nicely with my iPhone, in that I can go over my flashcards over the web using the Anki server.

Anyway, that’s the rationale behind how I approach it. Now the actual nuts and bolts of it.

For starters, I already had some college-level Japanese under my belt, but I had stopped taking classes about 3-4 years ago, then I stopped studying Japanese almost entirely for about 18 months. I had burned out, to be honest. Part of my frustration was centered around learning kanji. Ok, most of it. The other bits involved me being unable to do “output” properly. I couldn’t hold a conversation to save my life, and whenever I did, I felt very uncomfortable. I would always feel like I was unable to say what I wanted to say.

Remembering the Kanji

I had decided to start studying seriously again after 3-4 failed attempts to jump-start my Japanese. When I was looking for info on the web last summer, I came across James Heisig’s Remembering the Kanji. (125-page sample here.) Heisig came up with the system 30 years ago when he had to learn to be literate in Japanese fast. So he created one of the great quick-and-dirty methods to get a grasp on kanji.

His approach is simple. Break the process of learning to read and write kanji into discrete parts. He had noticed that native Chinese speakers had much less trouble dealing with kanji, because they already knew their own version from their language. They knew how to read and write them, and they knew what they meant in their own language.

So merely substituting Japanese readings/meanings was a lot easier.

It’s sort of like recognizing the letters of the alphabet for Latin alphabet-based languages. Even if there are a few squiggles or dots added, I can still see the ‘a’ in ‘ä.’ But alphabet characters don’t have individual meanings, so we don’t need to ‘name’ them, just memorize them. Also, 26 letters vs. 3,000+ kanji– you need a method of some sort!

Volume 1 teaches only the writing and one meaning in English for each character. It doesn’t teach you any Japanese at all. I’ll admit I was skeptical at first.

What good would it do to just know a vague meaning and how to write the characters?

I decided to bite the bullet anyway and try the sample chapters, and read his explanations for how his system worked. And you know what? I tore through the first 125 pages in a few days, and ordered the book from Amazon.

As I was doing it, I got it. I finally found something to help me tame the kanji dragon.

It’s actually a brilliant approach for Western language types, because it gives Western students a similar advantage that Chinese language speakers have in learning Japanese.

The ‘meaning’ you get it really just a label. Don’t think you’re learning the real meaning of the kanji! You’re not. You’re just slapping a convenient label on it so you can remember it later on. Remember, these are “English kanji,” therefore they have different meanings than the Japanese kanji you’re going to learn later on.

Don’t get hung up over whether you’re learning the real meaning. It doesn’t matter. The English keyword is made to be forgotten when you’re fluent.

You’re Ready to Move On!

So when you move on to either RTK volume 2 or some other method of learning kanji readings, you’ll have the hard part down. If I can peg a single English word with a basic concept to a kanji character, then I have a place in my brain to put it until I can successfully attach a Japanese meaning/concept/pronunciation to it.

As you learn real Japanese, the keywords fade over time.

RTK also banks on the power of the Division of Labor. Trying to memorize how to write, read, conceptualize, and pronounce (with sometimes up to 4-5 different ways to sound out a kanji depending on how it’s used) all at once is just begging to fail, because our brains aren’t that good at doing that sort of learning all at once.

It’s much easier to just learn a few things about a bunch of characters, get that down cold, then learn another thing, then another. It’s not impossible to do it the Hard Way, but I call it the Hard Way because that’s what it is.

While RTK volume 1 is a great book, and it takes a lot of work to finish, when you get there, you’re just going to be standing in the doorway. RTK volume 1 is merely the first preparatory step to learning one of the hardest parts of the language.

Sorry about that. But it’s true.

It sounds frustrating, but in the long run, it will make getting to the top floor a lot easier. Think of it as an elevator to the top of kanji mastery in comparison to grinding up the stairs.

The book isn’t easy to get through. It takes focus, patience, and time. When I first started using the book, I made a few tactical errors:

  1. I made paper flashcards. Time-consuming and inefficient. They took away from time I could have used to create vivid stories.
  2. When I finished making all of my flashcards, I took a break and stopped reviewing. Well, to be honest, I finished all of the flashcards, but I hadn’t even bothered to review the last 900 or so, because I had fallen behind on my reviews of the previous 1100. And so the downward spiral began, and then I started missing days, and that was that.
  3. I didn’t use a computer-based SRS to time my reviews out for me. I had the shoe boxes. Ugh.

So what’s a better way to do it? Easy. Spaced Repetition Software. I like to use the Reviewing the Kanji website to do my RTK reviews, but you could use any SRS program out there, like the above-mentioned Anki. (It comes with an RTK deck.)

The Reviewing the Kanji website is great because it has a shared database of stories its users have used to memorize individual kanji. The stories alone are a great reason, but there’s also a lively community on the message boards there too. You can find a lot of people to bounce ideas off of, not just for finishing RTK1, but for Japanese language learning in general. The site also has a solid SRS system you can use efficiently. I still use it for my daily reviews of RTK1.

On my second try in January, I plowed through the book in two and a half months.

Just one thing to keep in mind– it’s all about momentum. You create momentum, so you must maintain it. You don’t have to learn new things every day, you just have to keep up with your reviews. Don’t miss days. The more you miss, the deeper the hole you dig for yourself. And eventually, it will get so deep that you quit and wind up back at square one. Not a good place to be.

Links:

On theories of Language Acquisition:

Practical Applications in the real world:

Remembering the Kanji:

Remembering Stuff:

Interesting Places to Argue About this stuff:

To be continued in Part Two…

This blog is protected by Dave\'s Spam Karma 2: 3159 Spams eaten and counting...