kanji koohii FORUM
Am I actually learning? - Printable Version

+- kanji koohii FORUM (http://forum.koohii.com)
+-- Forum: Learning Japanese (http://forum.koohii.com/forum-4.html)
+--- Forum: The Japanese language (http://forum.koohii.com/forum-10.html)
+--- Thread: Am I actually learning? (/thread-5300.html)

Am I actually learning? - somstuff - 2012-07-17

I'm going through through some sentences is Anki right now (Tae Kim), but sometimes I feel like I'm not even learning the kanji. After seeing only part of the sentence, I recognize it and know it's full English meaning. From there, I think of how you say it. So, in my head I do full sentence>meaning>reading, as opposed to specific kanji>meaning/reading.

Another problem I have is that when I'm doing vocab in Anki (Genki, Core, Tae Kim), I feel like I'm only looking at one certain unique part of the kanji, and determining the meaning from that, especially for kanji I haven't yet done in RTK. That "uniqueness" is only unique based only on the vocab I've seen already. For instance, if I see 仕事, I really just look at 仕 and know that it is means "work/job" because it's really the only time I've seen 仕 so far. I understand that over time I will see more compounds with 仕, but idk.. it just feels wrong.

In short, I am looking at only parts of kanji or parts surrounding that kanji to determine its meaning, and I feel like I'm not completely learning it.

Am I actually learning? - Fillanzea - 2012-07-17

I think that this is kind of inevitable if you're doing something like the sentence method, as a beginner, and I don't actually think it's a bad thing. The more progress you make, the more sentences you'll see, the more you won't be able to rely on memorization, but for now this is the good kind of shortcut -- like training wheels on your bike, or having your parents read books to you while tracing along the text with their fingers.

If it really bugs you, then change your cards so that they test what you want to test for, whether that's individual words or individual kanji -- but I think the context is a good thing and I think you're learning more than you think.

Am I actually learning? - howtwosavealif3 - 2012-07-17

if I were you I wouldn't pay any mind to this. I wouldn't fret over this anki deck crap like making yourself choose fail instead of 2 or 3 because of this one aspect or going out your way to change the format or make another deck blahblah. you just have to get used to Japanese and for that you need immersion and time. For example reading more japanese which eventually leads to you getting used to reading japanese... for now I can recommend japanese song lyrics. It sure worked for me and it was fun... I used it to leran to read kanji and kana more smoothly/faster/etc. Just intall rikai-chan and getsome lyrics to asong you love and go crazy.

Am I actually learning? - CerpinTaxt - 2012-07-17

somstuff Wrote:Another problem I have is that when I'm doing vocab in Anki (Genki, Core, Tae Kim), I feel like I'm only looking at one certain unique part of the kanji, and determining the meaning from that, especially for kanji I haven't yet done in RTK. That "uniqueness" is only unique based only on the vocab I've seen already. For instance, if I see 仕事, I really just look at 仕 and know that it is means "work/job" because it's really the only time I've seen 仕 so far. I understand that over time I will see more compounds with 仕, but idk.. it just feels wrong.
To be honest the same thing happens to me in English. Sometimes i only need to see a part of a word to read it completely because you already know what the word is going to spell. It's one of the reasons I don't like reading aloud because I trip up sometimes. Or maybe this happens to just me who knows.

In any case I don't think it's a huge problem to be doing this. Pattern recognition is something that your brain picks up on easily and this seems like a classic example of it.

Am I actually learning? - Sequa - 2012-07-17

I started studying with sentences but had a similar problem like you and felt like I didn't learn much. It somehow feels like you are cheating and you get the meanings and readings right without really learning the words. I found that studying from kanji->meaning only really helped reading but was pretty useless for listening comprehension because once there were no Kanji, I could not understand anything.

Then somebody on this forum gave the advice to study from reading+audio to kanji+meaning+sentence+sentence-audio and I found that works a million times better.
If you can't look at the Kanji and don't have the sentence there is less opportunity to "cheat". Also, by having the word read out to you, it's less effort so it speeds up your reviews and is less boring so you're more likely to keep going. I seem to pick up the Kanji readings anyway so I don't think that's a problem but maybe that's because I knew many readings before I started this method. There is a disadvantage in that there are many homonyms but I can put up with that.

If you are like me and think listening is the most important skill (I prefer watching drama over reading books and for conversations of course listening comprehension is crucial. No Kanji to help you out there), then you should consider this.

Btw, I think putting the sentence on the front of the card is a really really bad idea in general. The sentence should only give you a context so you know how to use the word if you are unsure about it. But really it is the vocab word that you are trying to learn and having the sentence on the front really is cheating I think because it gives away way too much information.

Am I actually learning? - SammyB - 2012-07-17

Sequa Wrote:Btw, I think putting the sentence on the front of the card is a really really bad idea in general. The sentence should only give you a context so you know how to use the word if you are unsure about it. But really it is the vocab word that you are trying to learn and having the sentence on the front really is cheating I think because it gives away way too much information.
It's hardly cheating... When you encounter these words "in the wild" are they going to be a sentence or all by themselves? Is it cheating if the sentence you see it in helps you remember what it means? o_O

Am I actually learning? - Sequa - 2012-07-18

SammyB Wrote:
Sequa Wrote:Btw, I think putting the sentence on the front of the card is a really really bad idea in general. The sentence should only give you a context so you know how to use the word if you are unsure about it. But really it is the vocab word that you are trying to learn and having the sentence on the front really is cheating I think because it gives away way too much information.
It's hardly cheating... When you encounter these words "in the wild" are they going to be a sentence or all by themselves? Is it cheating if the sentence you see it in helps you remember what it means? o_O
If you know the meaning of the word in question purely because you remember the sentence so very well after seeing it many times, then I think that is indeed cheating.
With sentences in the wild it is a different matter because you actually have to understand the sentence whereas in Anki, you can understand the word regardless of whether or not you understand the sentence but you start remembering it by patterns you see in the sentence. Even anything memorable about the sentence (e.g. very long, has an exclamation mark, etc) will start giving away the meaning because you recognize these patterns.

As with English, when I encounter an unknown word and I understand it by its context and you ask me again 5 minutes later what that word means, in isolation, most probably I wouldn't know. This is not studying.

Am I actually learning? - s0apgun - 2012-07-18

In CORE, vocab words show up many many times other than their own card. Just from the sheer frequency you will learn the meaning one way or another. Also, I feel like what you stated would be the case if one was being lazy when studying instead of really drilling the cards. I usually test myself on the meaning and reading of the bold vocab word within the context before I show the answer. If I didn't answer both of them correctly I'll fail the card, which then restarts the scheduling. I feel frequency and a good personal effort will teach you the vocab even in context. I think the trade-off is fine because having context is less dull when studying.

Am I actually learning? - somstuff - 2012-07-18

^ I agree with both of your posts, Sequa. Mostly.

I've decided that with more examples it will become the case that you won't be able to rely on these external things. And even though you may have hardly learned it the first time you saw it, it isn't completely foreign (well, yes it is lol, but...) and that any exposure helps.

And I, unlike so many people here, like to do the reverse cards (English translation / Japanese meaning > Japanese/kanji). Among other things, this forces you to know the specific strokes.

Also, I think that just kanji on front with meaning/reading on back is good. You go through cards quicker than sentences, it's less mentally taxing, and often times the context given by sentences isn't all that necessary. For a lot of nouns, it's just like 木 = き, tree, which is used in sentences in obvious ways.

I have another question. On AJATT, Khatz says he never intentionally "memorizes." But if there's a word he doesn't know, and he puts that word/sentence into Anki, does he only rely on repeated exposure to grasp the word? He talks about how lazy he is and I'm assuming he doesn't (mnemonically) link words to definitions.
-Should I stop trying to use mnemonics to memorize kanji meanings AND readings? Just let repetition do it's job - a lot less mental effort? Is that what most people do? It's really hard to associate multiple kanji you've never seen before (I'm not done w/ RTK) to an English meaning and seemingly arbitrary Japanese reading. Especially if there are multiple of those in a sentence, though I do try to only focus on one at a time so individual cards aren't so intense. Even if I do know the kanji in a compound, the meaning is not always evident.
-Would that mean to lots failed cards that build up?

Am I actually learning? - s0apgun - 2012-07-18

You get used to seeing these words over and over again. You'll learn the meaning even if it is a very vague understanding you have of it by seeing it in context repeatedly. The other thing is in Japanese is very vague in general and a lot of words don't have a concrete meaning and depending on the context the meaning can change.

Am I actually learning? - Zarxrax - 2012-07-18

I was just learning the sentences rather than the words, so I quit doing the sentence method and just switched to vocabulary cards. In some cases I put a short phrase rather than a single word. I still put a sentence on the back and just read it every time I review the card, which helps to ensure I fully understand the context in which the word is used. I find that I learn much better this way, and the reviews are also much quicker.

Am I actually learning? - kainzero - 2012-07-18

i mixed in vocabulary words with sentences in my decks and i still run into the same problem: sometimes i'll see a word and i won't know what it means even though i should.

at that point i just decide that i need more sentences in my deck. or just f-it and keep going and hope osmosis works, which sometimes it does.

Am I actually learning? - Oniichan - 2012-07-18

My 2 cents- I found that the greatest benefit of following Heisig's method was not that I could write kanji, but that they were stored in some form in my head. I felt "primed" after finishing book 1, ready to engage them in the wild so to speak. I finished the last half of the first 2000 kanji at a rate of about 100/day. It didn't matter that my recall was not perfect initially.

Perhaps you could stop worrying about sentences/context (in the "context" of anki studying :-p) and just cram as many useful vocabulary words, as quickly as possible so that you can get to the payoff (in the wild encounters: streets/shops/manga/tv/games/Japanese forums etc.) Think, the more words I've crammed, the more opportunities I have of recognizing words (even if only vaguely).

Take something like the core 10k list, set it up as follows: kanji>Translation/audio/reading/sentence(if you have one). When reviewing, look at the expression, say it aloud and think of its meaning, flip the card, listen to the audio and say the word aloud, grade yourself only on your ability to recall the meaning and move on. Lather, rinse, repeat.

Set up another card type too: Kanji>Translation/audio/reading/sentence (if you have one) and only grade yourself on the reading. Keep it simple by testing just one thing per card so that you can go as fast as possible. Don't be afraid to fail many cards.

Memories seem to me to be about connections; the more connections you have in relation to a single item of memory, the more you own it. You can study the reading, sound, kanji, sentence context, dictionary glosses for a word all in one go, but your progress will be slower, I think; and you'll miss many opportunities to encounter words (a 2nd time/3rd time/etc.)

Am I actually learning? - vix86 - 2012-07-18

Oniichan Wrote:Take something like the core 10k list, set it up as follows: kanji>Translation/audio/reading/sentence(if you have one). When reviewing, look at the expression, say it aloud and think of its meaning, flip the card, listen to the audio and say the word aloud, grade yourself only on your ability to recall the meaning and move on. Lather, rinse, repeat.

Set up another card type too: Kanji>Translation/audio/reading/sentence (if you have one) and only grade yourself on the reading. Keep it simple by testing just one thing per card so that you can go as fast as possible. Don't be afraid to fail many cards.
With Anki 2.0, its actually even easier to set your decks up for this kind of split. That way you don't have to do something special to know "Oh this is a meaning card" "Oh this is a reading card."

I do still wonder how people handle world-only vocab cards where a word can easily end up with 1) Slightly varied pronunciation based on context and 2) Different meaning.

Let me give you an example, this is from core.
彼は非常に分別のある人だ. - Here the reading is ふんべつ and means "prudent, sensible"
ゴミはきちんと分別してください - Here the reading is ぶんべつ and means something like "divide up, classify, separate"

Another,
All of these are read あがる as their dictionary form.
今日は仕事が早く上がったんだ。-　Means "to complete/finish"
私たちは２階に上がった。- Means to "rise, go up"
彼は人前だと上がってしまう。- Means "To get nervous, or stage fright."
冷めないうちにどうぞお上がり下さい。 Means "to eat or drink."

あがる has quite a few possible meanings as well, its kind of like English's "run."

These are just a few words. You'll run into plenty more like these. How do you set you cards up to handle for this? Or do you just sit there and try and think of all the possible meanings and pass it if you get one of them? If you add a hint what would that be and would you consider that cheating or no?

Am I actually learning? - Oniichan - 2012-07-18

vix86 Wrote:...These are just a few words. You'll run into plenty more like these. How do you set you cards up to handle for this? Or do you just sit there and try and think of all the possible meanings and pass it if you get one of them? If you add a hint what would that be and would you consider that cheating or no?
No, not cheating. I just paste a Google image into the 'front/expression' field when a situation like this arrives. Yeah, it gives away the meaning, but I don't worry about that. Volume and thus speed is my goal.

I guess what I was trying to say is that there is an alternative to using anki for the sake of mastery; and that alternative is to think of "reviewing" as "priming". I cram them in as quickly as practical with the hope of having more and more "ahh" moments when I encounter them again outside of anki in native materials etc. I'm counting on these experiences for providing me context, new connections/levels of understanding etc. It's Heisig's "divide and conquer" and "peg" approach.

I just put this out there because perhaps there are some like me who will benefit more from just racing through a decent list like KO2001, Core etc. in a couple months, then moving on to native material; as opposed to carefully setting up nice decks with pictures, sentences, dictionary glosses, lots of audio etc. and then spending the next year or 2 working through it. E.g. Core6K -> would require memorizing about 100 new words (200 cards) a day for 2 months. This might be too much, but you get the idea.

My advice would be to pick a frequency list* based on whatever material one is most interested in (newspapers/books/movies&TV/radio/manga), cram it using whatever method is most efficient for them (anki/paper flashcards/Iverson's list method/etc.), importing it into anki once they are finished so that they can maintain their "pegs" and, finally, jumping into native material.

In fact, this was common advice a couple of years ago. 1. Finish RTK, 2. KO2001 or Core 3. Native materials. Now, it seems that some people spend too much time on step 1, step 2 and maybe even between them (looking for the perfect list, method, deck etc.) Why not just fly through steps 1 and 2 then move on to something a little more interesting?

*There are plenty of lists and tools for generating them on this forum. Check out the learning resources stickies.

Anki is still great for studying native materials. I'm thinking of dorama decks made with Subs2SRS etc. I'm also assuming that a person has basic Japanese grammar under their belt.

Am I actually learning? - vix86 - 2012-07-18

Ah, ok. I see. I can't say I still agree with it, but I know quite a few people here use the priming method with some decent success. I think its probably paired with the amount of native exposure you are getting too though, priming might not work too well if you don't turn around and read/listen/watch stuff.

The Priming method (aka: 'throw it at the wall and see what sticks' method) is great if you don't want to spend lots of time on SRS-ing and just want to work with the material, but the trade off might come at having to look up lots of words to recall what that one word was. I'd be interested to see studies as to which method tends to turn out better in terms of speed of picking up vocab though.