Skip to content

What does it mean to “know” a word?

What do you get when you review a word with flashcards or a spaced repetition system?

When you can go through a list and give all the correct answers, what have you accomplished? Do you know the words or are you simply adept at reciting answers?

What does it mean to “know” a word? At a minimum, you should be able to understand it when it’s spoken and when you read it. Beyond that you must be able to use the word when speaking and writing.

And if all of that is true, what have you accomplished when you review a word? You’ve read it, but out of context. You haven’t heard it. You haven’t seen it written in context so you don’t know what context to write it in. Ditto for speaking it.

So once again, what have you accomplished?

And when you take a word in Japanese and look it up its english translation so that you can review it, what do you have? Can you use this new japanese word in Japanese as you would its english translations in English?

If not, what is the value of learning this direct translation?

I’m not suggesting that there is no value to translated meanings or to reviewing. I’m asking how much are they really worth? And given how easy it is to burn out reviewing vocabulary, I think it’s a very important question.

Flashcards: The Bane of Human Existence

How do you take something that’s fun to learn and make it soul crushingly boring? Make flashcards.

I’ve made flashcards by hand with my hideous handwriting (the technical term is “chicken scratch”). I’ve bought thousands of Kanji flashcards. Good cardstock, cheap cardstock. Curved corners and decks of paper cuts waiting to happen. All of them filled with all sorts of fantastic information about the kanji character - and even a few compounds it appears in!

All these cards did one thing and one thing only: They made me want to learn Japanese less.

Every card took something fun, learning about the Japanese language and culture, and reduced it to painful, boring, mind numbing repetition.

I still can’t believe that I ever thought that you had to memorize every On and Kun reading for a Kanji to be able to say you “know” it. You realize how many readings 上 has, right?

But you know what’s worse? Knowing every reading and every translated english meaning doesn’t mean you know that character. It means you’ve memorized a list of bits. It means you successfully forced the data into your head despite your brain’s best efforts to prevent it.

The average child falls about a 1000 times when learning to walk. If you want to stunt that learning process, sit the child down and attempt to explain the mechanics of walking.

But here’s the kicker: The child wasn’t learning to walk. He was just trying to go faster. If he can go faster, he can explore more, discover more, play more. The learning, the walking, came for free.

Flashcards are for learning. But true learning comes from exploration, discovery and play.

Japanese became fun for me when I stopped studying and started playing - The learning comes for free.

What Did Your Last Test Teach You?

I’ve never been a fan of tests or reviews for the simple reason that they are rarely used to move you forward.

Did poorly on a test because you had difficulty understanding the material? That’s nice, the class will be moving on now.

Reviews, Tests, Quizzes and the like are typically used to determine what you know. That’s backwards. Tests should be used to isolate what you don’t know.

What do you have difficulty grasping? What takes you longer to learn? If the objective is to obtain full understanding of some concept or idea, who cares what questions you got right? We need to work on what you got wrong.

This isn’t about getting cute little stars on your paper or currying favor with a professor because of your diligence - This is about learning.

And with Japanese there is a lot to learn and while most of it will be learned though listening and reading, we nudge ourselves forward by reviewing Kanji and Sentences.

But we don’t want the typical review. We don’t want to arrive at the end of the review with a correct list to be happy about and an incorrect list of more work to do.

If a review can determine what we don’t know, then it can help us learn it. A test can evaluate and teach.

When we only use a review to measure what we know, all we get is a snapshot of our “knowledge”. But if we use a review to measure what we don’t know we can leverage that information right now to move our “knowledge” forward.

Said another way, we should know more at the end of the review than we did at the beginning.

And when the review is done, all the reviewing is done. No additional studying, dictionary mining or grammar hacking. Just watch your JDrama, enjoy your anime, listen to your podcast, or read your book or manga - Because that’s where the real learning occurs.

Starting with Heisig

Keep it simple. It’s amazing to me how many people learning Japanese do everything within their power to not read anything in Japanese. As though there is this magical point when they have memorized enough Kanji, drilled enough vocab, and crammed enough grammar, reading fluently will spontaneously occur.

Now, to be fair, reading Japanese requires that you learn hiragana, katakana and Kanji. How many you need to learn is irrelevant - if you want to learn Japanese you have to learn Japanese.

But still, there are a lot of Kanji. How do you get them in your head? Some say, use Heisig’s “Remembering the Kanji” mnemonic/story techniques. I say no. My objection won’t be as colorful as Tae Kim’s but here we go:

  • It’s a form of work - not play
  • It’s easy to burn out
  • It uses too much English
  • Knowing Kanji is part of the fruit of knowing Japanese, not a prerequisite.

My first two objections are easy. Drilling of any kind is soul crushingly boring. While useful and in some ways invaluable, reviewing should be kept to an absolute minimum (I’ll talk more about this later). The idea that anything approaching a majority of my time should be used for recalling stories is repulsive to me.

My third objection has a little more meat. As a native English speaker, I have been and will continue to use English crutches to support my learning. But manufacturing english stories around each and every Kanji character is overkill. Especially, as Tae Kim points out, when you only view those Kanji outside the context of any compounds, phrases or sentences.

My last objection goes the deepest - to the danger of quantifying the unquantifiable - turning becoming fluent in Japanese into having knowledge of Japanese.

If you become fluent you will have lots of knowledge. But having lots of knowledge doesn’t beget becoming fluent.

You can see this quite clearly in the english sentences all over the internet written by people for whom English is a second language. They have tremendous knowledge of the language. Sometimes their knowledge surpasses even that of native speakers, and yet their speaking and writing belies their knowledge.

Fluency in a language is unquantifiable. You cannot reduce it to a number. You can observe people who are fluent and build a description of what they know but you cannot turn that into a prescription that someone else need only acquire to become fluent.

If someone who’s fluent in Japanese knows 2000 Kanji, 20000 vocabulary, and 500 grammar rules (or however many there are - I have no clue what it is for English…), then you have a description of what a fluent speaker has.

Memorizing those Kanji, vocab and rules will not make you fluent and I think Heisig’s technique encourages people to think the opposite.