I've finished learning about 200 onyomi pronounciations, which is more than enough for me at the time. I want to read stories now, but a lot of words I'm coming across involve kunyomi, so I thought I'd learn a fair amount of those next. But I'm finding that I'm having a very difficult time with this. It is terribly difficult for me to keep the kunyomi pronounciation straight from the onyomi pronounciation. When I am reviewing kunyomi vocabulary, I always end up pronouncing the onyomi instead, or the other way around. Are there any methods of remedying this?

By the way, I am not using RTK2. I am simply learning pronounciations by grouping similar compounds, which is easy if you look up kanji in compounds on the JLPT study page.

Kun yomi are almost always used alone, while on yomi are used in compounds. Thats the easiest way to keep it straight, except for the occasional exceptions, which I guess you have to learn on a word-by-word basis.
I would recommend not drill kun yomi and on yomi seperately... mix all your vocabulary together.

Also you might have much better luck using sentences rather than just drilling words. This gives you context, which you don't get from just reading a word by itself.

Are you learning the readings by kanji->on and kanji->kun? Because if you're using words that the readings appear in, it should be easier to know which reading is being used.
意見　いけん
見る　みる
I don't know how you could guess けんる for 見る.
For kun readings that don't have okurigana, like 火(ひ)　or 花(はな), you can tell them apart from on readings because single kanji words are usually kun readings not on.

Review words with both kun and on readings, in sentences rather than in vocab lists, and there should be little problem to learn them both together and distinguish them.

I dont even know why anyone would bother memorizing the kun/on readings. Like everyone says, memorize words in context. The only time I can even comprehend memorizing rote lists of readings would be if you're already fluent at SPOKEN japanese and just need to get literacy. Generally if you encounter a brand spankin' new word, even if you had a surefire way of divining the reading, that'd just transform it from an unknown compound to an unknown blur of sounds.

As an aside, you should be aware somewhere in the back of your mind of the dark black evil, conjured by Sauron to oppress us all, which is called rendaku (連濁). Rendaku is what makes 人々 be read "hitobito" instead of "hitohito", and there's absolutely no way to predict it. So, even if you memorize all the readings, rendaku will still force you to learn individual words

Wikipedia: http://en.wikipedia.org/wiki/Rendaku

snispilbor Wrote:I dont even know why anyone would bother memorizing the kun/on readings. Like everyone says, memorize words in context. The only time I can even comprehend memorizing rote lists of readings would be if you're already fluent at SPOKEN japanese and just need to get literacy. Generally if you encounter a brand spankin' new word, even if you had a surefire way of divining the reading, that'd just transform it from an unknown compound to an unknown blur of sounds.

As an aside, you should be aware somewhere in the back of your mind of the dark black evil, conjured by Sauron to oppress us all, which is called dakuten (濁点). Dakuten is what makes 人々 be read "hitobito" instead of "hitohito", and there's absolutely no way to predict it. So, even if you memorize all the readings, dakuten will still force you to learn individual words

Wikipedia: http://en.wikipedia.org/wiki/Dakuten

Actually it's called 連濁 and it can be predicted fairly well according to various rules. There are of course exceptions. Dakuten is just the way in which rendaku is expressed in written (kana) form.

http://en.wikipedia.org/wiki/Rendaku

I do like the idea of learning words in context, but there is a small problem.

I have a hard time remembering the word and readings for that particular sentence. Because of this, I just have to keep failing the sentence over and over again until I am able to learn the word. So I now have a vocabulary list for Anki and then a sentence list with those vocabulary words in it. It seems to be working okay, but I am still having a VERY hard time to learn the words, and the review takes so long I don't have enough time in a regular day to add new sentences.

So it's these new words I learn that really get me. Is there really no better way to nail new vocabulary into my brain? Any tricks for remembering readings, at all? I have tried associating the reading of a Kanji with its story, but since the story is already made I can't remember any additions I've made to it for the readings...

Have you just started studying Japanese?

Try learning words in hiragana first, and then test yourself by reading the hiragana and trying to remember and write the kanji.

If you've just begun, make sure you have the hiragana under your belt, or else it could make recalling readings harder.

Even though it's essential and necessary to learn kanji, and having done RtK it is even more alluring to make sure you learn every new word with its respective kanji writing, if you are really just starting, it might be good to practice things in hiragana for awhile until you can read things in hiragana with little effort. Throw a few easy kanji words into the mix because all hiragana text can actually be harder to read sometimes, but until you've mastered the phonetic parts that the kanji represent, it might be hard to recall them from just seeing kanji.

Don't try associating readings with stories. It's really just a lot of practice I guess and reading things over and over. What do you do now to learn the words, what level are you, and how many new words are you trying to learn in what kind of amount of time?

AryllWind Wrote:I do like the idea of learning words in context, but there is a small problem.

I have a hard time remembering the word and readings for that particular sentence. Because of this, I just have to keep failing the sentence over and over again until I am able to learn the word.

Do you have the readings of the word on the "answer" side of the card? That's what I do.

Another thing I do is. Whenever I add a new word. I add a BUNCH of sentences that use that word. Paradoxically, the MORE sentences with the new word, the FASTER it seems to be to understand the whole pack. For instance if I'm forced to use just 1 or 2 sentences for an obscure word.. then those sentences end up bouncing back and forth between "missed" and "barely remembered" for weeks! But if I get like a dozen or more sentences, I practically pass them all in a day.

Of course the hard part of this is... when you first start out.. it's hard to find a sentence using your word, that doesn't use other words you don't know yet. It might be worthwhile to learn a core 1000 common words the old fashioned way (rote memorization). If you do this, and want to avoid English, for very concrete nouns you can use pictures instead of English translations.

When I was just starting out, I had a much harder time memorizing new words. Now that my SRS has grown to over 6000 cards, I seem to have a better "feel" for japanese and even when a word is totally new (no familiar rootwords) I can learn it very fast now.

Speaking of rootwords... I just figured this technique out... if a word is long and/or you're having trouble memorizing it.. try seeing if you can identify any rootwords and then add other related words/compounds/sentences using those same roots...

EDIT: Here's another thing about vocab... I watch about 3 hours of Japanese youtube content per week... and have been doing so for about a year... and with just that little bit of input... I'm ALWAYS realizing there are words in my lexicon that aren't in my SRS. Just last night 気分 (kibun) showed up in an example sentence and I realized I knew it even though I'd never formally focused my study on it!

AryllWind Wrote:So it's these new words I learn that really get me. Is there really no better way to nail new vocabulary into my brain? Any tricks for remembering readings, at all? I have tried associating the reading of a Kanji with its story, but since the story is already made I can't remember any additions I've made to it for the readings...

Well, for kunyomi, at least, I use a method that looks a lot like what Heisig did for the writing. I use (ON and KUN) readings I already know, as well as words, as "primitives" from which to build the rest of vocabulary. With these primitives, I build up stories that links meaning with pronounciation. The good thing is that I'm free of using any "primitive" that has the reading I need. Two examples:

1. 祈る means "PRAY" and is read いのる. I pick 椅 (イ, chair) and 乗る (の, to get on). Then I build my story: In order to pray, the monk gets on a chair.

2. 甦る means "RESUCITATE" and is read よみがえる. For this I pick 黄泉 (よみ, Hades) and 帰る (かえる, to return). Here note that I allow か when the change to が makes sense. The story: Resucitate is to return from the Hades.

These mnemonics are a bit less useful in that, once you become familiar with a word (which doesn't take much), you stop using them, but are great for adquiring vocabulary. The good news is that, by reviewing a word, you also review the primitives from which it's composed.

There are some caveats, though.
1. Even though I review from text to pronounciation, I try to avoid using the primitives in the kanji with my new story, to avoid contamination of my pronounciation mnemonics with my writing mnemonics.
2. There are multiple KUN readings for some kanji, and in some cases there are multiple readings for the same combo of Kanji+nanori (怒る for example, can be read as いかる or おこる).
3. Restricting to JLPT-level vocabulary lists for learning Kunyomi looks as bad as learning kanji on traditional order, but learning every word is infeasible. My approach is to restrict myself to single kanji words and compounds that got irregular lectures "glued" onto them (like 黄泉), but it's just a first thought.
4. I don't have the kanji ordered in a progressive fashion that would optimize the learning of the KUN readings. I don't know if Trinity handles this, or if there's a book like Heisig's for mastering KUN readings, in other case, somebody could start writing one. Ideally, I would settle for which "primitives" are best for a given word and, once all primitives were fixated, an order could be chosen that worked best.