RECENT TOPICS » View all
うるさいよ....日本語を勉強しましょう。
Last edited by taijuando (2008 June 19, 3:08 am)
What's scary is that it seems refined down to almost a science. I honestly think I can take my spreadsheet of sentences and get any one from Japan or the US speaking Japanese or English in a reasonable amount of time.
The Japanese people I've shown the English variant (Anki program, UBJG english sentences, TTS audio) were very excited. The sailors I showed a vocal variant only of Japanese (no kanji, just TTS and kana) were also excited. That and they just listen/watch their target language all the time. Then I tell them to google the images of the words targeted in the sentence for better impact. It's just hard to convince either group the real benefits that RevTK can offer.
So yeah, if Khatz has little substance to add, perhaps its time he shows off. Let's face it, in just a short year the way many of us learn Japanese has been turned on its head. The next year will bring in more to the fold. That means many people posting about their various successes along the way. Khatz is just leading the pack with his own success.
I wanted to ask you guys your view on something. The AJATT method starts with RTK Vol 1 only, and then moving on to sentences etc. In RTK, Heisig is recommending that the study order should be keyword -> kanji, but I'm starting to wonder if this is such a good idea when doing AJATT. Isn't the whole point of doing RTK Vol 1 to learn to recognize the kanji and their meanings on sight? Doing english -> kanji seems like the wrong order to me. Sure, you get to practice writing the kanji, but it seems harder to actually remember the characters and that has also been my experience so far since starting RTK. I can do the keyword into the kanji without much trouble, but if I'm actually reading some Japanese and come across the kanji I've studied, I have problems remembering their meanings. What's you guys' take on this and the role of RTK1 in the AJATT method?
Shirow66 wrote:
I can do the keyword into the kanji without much trouble, but if I'm actually reading some Japanese and come across the kanji I've studied, I have problems remembering their meanings. What's you guys' take on this and the role of RTK1 in the AJATT method?
Don't worry. This only happens a few times for each kanji with me, and mostly with the newly learned ones.
There's a thread in this forum about this:
http://forum.koohii.com/viewtopic.php?id=1420
Sarius24 wrote:
I would like to just say something about the 24/7 listening to Japanese. If you are wearing headphones 24/7 or more than 5 hours with very high music you'll be damaging your ears so I recommend you burn your Japanese music to some cds and play em on a stereo. Headphones can be dangerous so keep your hearing healthy turn down your music if you plan to listen to lots of Japanese.
Arigatou gozaimasu.
While you should be careful about damaging your hearing, there is no difference between playing headphones too loud and playing speakers too loud. Just play as quiet as you can while still being able to hear the Japanese ![]()
uberstuber wrote:
While you should be careful about damaging your hearing, there is no difference between playing headphones too loud and playing speakers too loud. Just play as quiet as you can while still being able to hear the Japanese
The problem is with background noise. If environmental noise is present, then the volume is higher than you think it is. If you've ever driven at highway speeds while listening to the radio, and then come to a stop at a quiet intersection, you've probably noticed how much louder the music suddenly seems.
That's the real reason everyone's getting hearing loss from portable music players. It's not an intrinsic property of the iPod or the earbuds; it's that people aren't listening in quiet environments.
I'm getting the impression that it is unavoidable that the AJATT method (or sentence method) results in an excellent command of Japanese grammar.
I got this feeling when I saw the word 'grammar': 文法 (ぶんぽう), which in Heisig keywords gives 'sentence method'. ![]()
I can assure you that it is entirely avoidable. I base this on my observation of the English grammar of certain native English speakers I know.
~J
Hmmm... but probably they didn't learn their English via the AJATT method and a SRS? I'm sure that must be the explanation. ![]()
abaddon wrote:
Hmmm... but probably they didn't learn their English via the AJATT method and a SRS? I'm sure that must be the explanation.
Hmm, perhaps it's time for 'Remembering The Roman Letters'?
'T looks like て'...
Last edited by stshores24 (2008 July 12, 10:32 am)
woodwojr wrote:
I can assure you that it is entirely avoidable. I base this on my observation of the English grammar of certain native English speakers I know.
I'd say that by definition, they speak their language perfectly, since it's by their speaking it that the language exists. A language is defined by its native speakers. It's simply the grammar as a descriptive tool that's at fault.
Hi guys.
I recently completed RTK1 (which is to say I finished entering them but still have many reviews each day) and have started on AJATT. I am looking for a bit of feedback on my methadology which is as far as I can tell the latest method. It should be noted that I have been living in Japan for about 1.4 months and studied for a year before coming to Japan, completely avoiding the kanji. So I have a pretty good vocab and understanding of grammar. So at this point my main goal is to bring my knowledge of kanji up to par with my spoken Japanese.
I am using KO (Kanji Odyssy) and entering sentences into Anki. The question is the Japanese spoken and the answer is the sentence written in kanji, including explanations of any words that are new to me. I am not reviewing answer to question.
How do you guys do it? Do you do it both Q to A (production?) and A to Q (recognition?)?
I am also curious how you actually remember a new word. Take 提供 for example, which refers to some kind of offer or program on offer. From the Heisig meanings we have propose and submit. Do you make a story using the part that leads to the kanji meaning, like the methods used to learn the individual kanji?
I am trying avoiding this whereever possible, since I don't want English to be mixed into my Japanese anymore than it is (which I grudginly accepted doing given how fast the Heisig method is). I will however, if I continually fail to remember some kanji, make a picture in my head combining the elements, but avoid using avoid using any story, at least in English.
Anyway, I have read most of this thread, but didn't find much incorporating the flip method (http://www.alljapaneseallthetime.com/bl … panese-too). I am thinking since I am just starting, now is the time to iron out the process.
How do you guys do it?
My style for anki:
If I have words I want to learn how to write, I utilize two cards with flip sides (Ie, Q->A and A->Q). One model will quiz reading, the other will quiz writing.
Sometimes if the sentence has lots of unknown words/writings (say, 4+), I will make it so both models are like 1/2 Q -> 1/2 A. Instead of taking 4 words I want to learn to write and put them all on the Q side as Hiragana, I will put half of the words in Hiragana on the Q side (with their kanji on A side) and then the other half with Hiragana on A side (kanji on Q side). This works very well with card spacing. This might have been confusing soo.....
Example from my deck (Final Fantasy 7):
Front:
「ここから先は立ち入りきんしだ。
俺はここでおまえみたいなふろうしゃを
追っ払うケチな仕事をしてるもんだ」
「わかったら、さっさと失せる!!」
Back:
「ここから先は立ちいり禁止だ。
俺はここでおまえみたいな浮浪者を
おっぱらうケチな仕事をしてるもんだ」
「わかったら、さっさとうせろ!!」
So both times I see the card, some word will be tested on how to write.
(I should also note I stopped putting entire cards in Hiragana. I am only putting words I want to test into hiragana. I do think, however, you should test yourselves on simple words even. So while you don't need a card entirely in hiragana, 学校 OR 明日 should be in hiragana somewhere in your deck for you test unless you're really, really, good at writing the word already.)
Sometimes (often?) I use heisig meanings to remember vocab. I don't think it will have a negative effect -- especially if you are using an SRS. I find that although I might need to use English keywords to help remember writings for awhile, after some time the SRS makes the writing (and speaking of a sentence) become very natural.
Last edited by sutebun (2008 July 19, 12:54 am)
thermal wrote:
Hi guys.
How do you guys do it? Do you do it both Q to A (production?) and A to Q (recognition?)?
I am also curious how you actually remember a new word. Take 提供 for example, which refers to some kind of offer or program on offer. From the Heisig meanings we have propose and submit. Do you make a story using the part that leads to the kanji meaning, like the methods used to learn the individual kanji?
How do you guys do it?
Due to a recent post on AJATT, I've been doing double cards: Audio (with kana if necessary) to Kanji Sentence - I hear the sentence then must write it out in kanji. Kanji Sentence to Reading - I see the kanji sentence and must read and understand it entirely, compared to the audio/kana/definitions.
For individual words, you can just put them in there own flash card set called "vocabulary". I've been using Trinity for this recently. Use any method that works when it comes to memorizing them though. It can be Heisig phoenetic pictures (explained in the Kun yomi portion of RTK2), keyword game, other mnemonic, brute force, etc.
I have a couple of questions about RTK + the AJATT method that I hope you can help me out with.
1) I understand that the AJATT method states that you should complete Remembering the Kanji/Kana first and then move on to sentences. Is there any benefit or detriment to working with sentences before you are done RTK? Has anyone done this?
2) Regarding RTK with Anki (or another SRS), Heisig seems to want you to go from Keyword -> Kanji. This seems like the most logical way to get you to remember how to form each Kanji but I know others have stated that sometimes they have trouble identifying Kanji when reading because they aren't used to going from Kanji -> Keyword/meaning. When using the AJATT method is it better to go from keyword -> Kanji or from Kanji -> Keyword?
3) Sentences are a big stumbling block for me. I understand the concept behind it and it seems like a solid way to learn but I just can't wrap my head around the execution, especially for people who know very little vocab or grammar.
Assuming I use Kanji Odyssey, I'll have a sentence and it's English meaning. Both go into a SRS. When reviewing I copy the sentence then write out the readings of each Kanji and then try to understand the meaning of the sentence? When starting out you are going to basically be converting the Japanese sentence to some sort of English meaning but eventually you'd want to move more towards understanding fully in Japanese? Is that correct?
How does one go about trying to understand grammar when beginning with this method? I'd think that you'd be so overwhelmed with vocabulary and meanings of Kanji that grammar would almost be an afterthought. Even if it's not you'd have no real starting point for your grammatical study since sentences could contain any number of grammatical constructions. Am I missing something here? I feel like understanding this is the key to unlocking full understanding of the AJATT method for me.
1) RTK is achieved very quickly. Some people can do it in a month. I'll finish probably in 2 days, closing allmost 4 months with 1h/day study.
I'm not working in sentences yet because after I'm done with my daily RTK digestion I have not the time or I am very tired.
Doing one thing at a time is more time efficient, as RTK seems to gain its full power only after its end. That's because it saves some of the more coomon kanji as desart in the end of the book.
Doing the kana after kanji is kinda good idea. I never drilled hiragana, as it came to me so easily after just one pass trought "remembering the kana I". I intent to redrill them after I finnish RTK, as it seems there will be so little stress, and I need a break from RTK.
2) People seem to do it with great results, but only after finnishing RTK. Doing it first is counter productive and time wasteful.
3) Start with the simplest sentences. Evolve from that. Grammar is very useful after you understand it. But to understand grammar you need to have acquired its usage.
Katzumoto says it like it is something new, but talking with my mother, that is a teacher, she pointed out it is old stuff, and is used all over the world to teach small childreen how to read:
http://en.wikipedia.org/wiki/Constructivism_(learning_theory)
Last edited by mentat_kgs (2008 July 19, 3:06 pm)
kobeblackcat wrote:
I have a couple of questions about RTK + the AJATT method that I hope you can help me out with.
1) I understand that the AJATT method states that you should complete Remembering the Kanji/Kana first and then move on to sentences. Is there any benefit or detriment to working with sentences before you are done RTK? Has anyone done this?.
I tried the sentence method while at the 1000 mark in RTK. Not very good success with it. You're getting sentences that have kanji that you don't know yet. Which means you're learning only grammar and not reading. The purpose of the sentence method was to get your reading up to snuff (amoung many other things). So yeah, best to finish RTK first. That said, doing an audio course (say like Pimsleur) and listening to Japanese all the time while doing RTK is really beneficial.
kobeblackcat wrote:
2) Regarding RTK with Anki (or another SRS), Heisig seems to want you to go from Keyword -> Kanji. This seems like the most logical way to get you to remember how to form each Kanji but I know others have stated that sometimes they have trouble identifying Kanji when reading because they aren't used to going from Kanji -> Keyword/meaning. When using the AJATT method is it better to go from keyword -> Kanji or from Kanji -> Keyword?
.
I really like RevTK, but it does not give you the option on this site to do Kanji to Keyword while tracking your keyword to kanji. Using an SRS, there's no harm in doing both kanji to keyword and keyword to kanji. You'll find your kanji to keywords pass at a higher rate (thereby getting reviewed less and not interferering with your keyword to kanji. In addition, you'll find those odd kanji where the reverse is more difficult.
Now if you do ONLY one thing, it must be keyword to kanji. If you do both, it MUST be in an SRS that tracks both (so that you don't see both cards too close together) for the best benefit.
kobeblackcat wrote:
Assuming I use Kanji Odyssey, I'll have a sentence and it's English meaning. Both go into a SRS. When reviewing I copy the sentence then write out the readings of each Kanji and then try to understand the meaning of the sentence? When starting out you are going to basically be converting the Japanese sentence to some sort of English meaning but eventually you'd want to move more towards understanding fully in Japanese? Is that correct?
How does one go about trying to understand grammar when beginning with this method? I'd think that you'd be so overwhelmed with vocabulary and meanings of Kanji that grammar would almost be an afterthought. Even if it's not you'd have no real starting point for your grammatical study since sentences could contain any number of grammatical constructions. Am I missing something here? I feel like understanding this is the key to unlocking full understanding of the AJATT method for me.
Don't think of it as converting Japanese to English. You want to understand the concept of the sentence and each word in the sentence. The English is there to help you get the concept of the Japanese quicker, not to be the ultimate answer.
As for grammar, start with sentences from a textbook (I recommend UBJG) that slowly build on principles. In addition, you can read Tae Kim's site to get an idea what it is you're seeing. From the simple comes the complex in time. Hopefully with full literacy.
Nukemarine wrote:
Due to a recent post on AJATT, I've been doing double cards: Audio (with kana if necessary) to Kanji Sentence - I hear the sentence then must write it out in kanji. Kanji Sentence to Reading - I see the kanji sentence and must read and understand it entirely, compared to the audio/kana/definitions.
Do you have a link to this post? I had a look at recent posts but couldn't see anything like this.
I looked for that post as well and couldn't fine it.
If I have one complaint about AJATT it's that the site isn't all that user friendly.
thermal wrote:
Nukemarine wrote:
Due to a recent post on AJATT, I've been doing double cards: Audio (with kana if necessary) to Kanji Sentence - I hear the sentence then must write it out in kanji. Kanji Sentence to Reading - I see the kanji sentence and must read and understand it entirely, compared to the audio/kana/definitions.
Do you have a link to this post? I had a look at recent posts but couldn't see anything like this.
http://www.alljapaneseallthetime.com/bl … y-calendar
In addition, the AJATT thread on this forum also talks about it. Just look around the May 4th, 2008 timeframe.
I think we're ultimately creating what Rosetta Stone tried to do but fail: Create Audio/Visual cues to deepen the learning/retention of the language we're studying.
We have: Sentences we care about, extra info we need, photos if we deem appropriate, english, kanji and kana variants, audio of the sentences, and most important S to the R to the freaking S!!!!
Thanks Nukemarine. It doesn't really say to do the facts as both production and recognition as far as I can tell.
I actually started thinking about the words and realised that I do remember the reading of pretty much all the new words I have learnt doing KO. I have also found a good book with some graded reading which is really ideal. It's called Enjoyable task reading in Japanese: Pre-Intermediate, which has reading tasks at a nicely graded level. Some are just reading comprehension and some a logic/math puzzles.
I also read new sentences in KO without looking at the English or furigana (utilizing two advanced pieces of cardboard) as another aid.
I think I will stick with just production for the time being and see if it doesn't all fall in a heap ![]()
Someone (probably Khatzumoto) should put up an "AJATT FAQ" section. It would super useful ![]()
EDIT: I know, it's a very lazy post and I could do it myself... I'm just lazy today, I'll think about it another day ![]()
Last edited by nac_est (2008 July 20, 9:52 am)
I haven't read the whole topic so I guess my opinion on AJATT has allready been stated by others, but here we go.
Personally, I really don't see what people like about it. The difference between common sense japanese learning (learning kanji, grammar and use japanese) and AJATT (use japanese until you somehow understand it) is that AJATT is just a small part of normal learning, with important parts excluded.
Anyone serious about learning japanese knows that you can't simply study grammar and vocab lists, you eventually have to actually read real japanese texts and write your own japanese texts etc. The AJATT method sounds more like "screw tedious grammar lessons and such, just read japanese texts until you understand!". Sounds like a magical cure when in fact it isn't.
My personal opinion is, study grammar because it helps a LOT. It doesn't take a lot of time to learn all the grammar you need to start reading japanese novels, watch japanese movies etc. Then, when you actually know some japanese, it's a great idea to start surrounding yourself with japanese stuff. Personally I read japanese novels, japanese manga, i watch Jdramas, animes and movies in Japanese, I play japanese games, I speak nothing but japanese with my girlfriend, I speak nothing but japanese with my neighbours.
Summary: Use as much japanese as you can, but don't neglect actually studying grammar.
SRS is just a super efficient revision method which pretty much allows you to remember everything you've ever studied. So the method should be study grammar first, then load up your SRS with said grammar so as not to forget it. Same with vocab, study first, revise second, and use third.
Production versus recognition is a really old language question. You have two roads, have a large passive vocab with a small active part, or have a small vocab which is almost completely active. Even natives have a very much larger passive vocab than an active one, so personally I feel the need to understand more than I can say. So I do recognition only, at a faster rate, and it slowly trickles through as I actively use it, or as I cram in more passively.
Also, it should be said that at first you will know literally nothing. So in the beginning stages it may pay off more to cram hard and fast, and later to try to use it a lot, instead of trying to use the tiny amount you know from the beginning. A bit like Antimoon's idea, passive before active. Get used a lot of structures and have a late start in producing.
Part of the problem with the "I can this later, I can do that later" is that you forget that later has come and you've not done it. If you start off always playing Japanese Music, Always watching Japanese shows, always have a Japanese operating system, always have Japanese posters, always eat Japanese food, etc., etc. is you don't have to worry when later comes and you can begin to understand what's surrounding. You're already surrounded.
Trust me, I live in Japan and I have to GO OUT OF MY WAY just to get in a Japanese environment. I hear all the excuses in the world about what people will be doing later. And later comes, they still haven't done it. I'm the weirdo cause I am doing it (well, and for the way I exercise in the gym).
And I think you misunderstand the 'screw grammar'. Grammar is VERY important in AJATT, it just becomes a natural byproduct is all. I'm not thinking, "hmm, to I desire to do something but am unable to, I must use a Negative base followed with a Volitional base and tai." What has happened is I've experienced enough sentences with the desire to do something but being unable to that it just seems right.
Yeah, I don't subscribe to the everything that's suggested in AJATT (for one, I'm not the type to watch the same thing over, and over, and over). However, I am noticing some misunderstanding from time to time of what is being suggested.

