I did not use RTK. I think it is a waste of time. I straight up used flashcards to memorize on and kun readings for my first 1000 kanji, and after realizing how tedious and inefficient that method was I decided to just learn new kanji along with new words.

You may wanna check out Textfugu tho.
I just can't seem to stick with RtK.
But it wasn't a waste of time either.
RtK made me familiar with the shapes of Kanji and
all future learning in this regard was greatly eased.

I seem to be the type that better learns the Kanji in context and/or with vocab.

Revenant Wrote:You may wanna check out Textfugu tho.
I just can't seem to stick with RtK.
But it wasn't a waste of time either.
RtK made me familiar with the shapes of Kanji and
all future learning in this regard was greatly eased.

I seem to be the type that better learns the Kanji in context and/or with vocab.

I have looked at Textfugu, and it is inconsistent with my goals. I know that many people will find it useful. One of my goals is to learn more about the history of specific Kanji, and how they evolved over time.

I think I fall firmly in the middle of the 'to use RTK or not to use RTK' debate.

I started out very enthusiastically working my way through, faithfully memorising and using Anki to help with this. I got quite far into the book - about 1600 as I recall. This took me about 3 or 4 months of doing nothing but RTK.

I found this very hard going, and although I was able to easily recall the meaning of the characters it did not help me decipher meanings, and of course it didn't help me read. I decided something must be done - if you spend 4 months studying something and it's not helping you achieve, it's time to really consider what your aims are.

Since my aim was to be actually able to read something I put RTK aside and started to actually read stuff. For a while I tried extensive reading - where I read really simple texts without a dictionary. Then I started reading blogs and gossip pages, and later on to magazines I picked up for cheap off ebay (I paid about 3 quid per magazine). I do like extensive reading without a dictionary for the sheer thrill of 'reading a whole book' and understanding almost everything, even if I can't define each word. Reading blogs and gossip pages, as well as manga as helped me pick up common words and phrases.

In conclusion, RTK has helped me to memorise how to write kanji (so when I see a new word, I sometimes think of it in terms of the key words - like 可愛い I would memorise as 'can love'). However, it's purpose is clearly not to learn to read, and I think if your aim is to learn to read, RTK should form the backbone to varied and extensive reading of native materials. I honestly think that people who finish RTK before reading anything are either very smart, or like to punish themselves!

RTK is supposed to be the basics and the backbone to learning how to read and write through whatever means you like. You get the hardest part over with first, which is the remembering, recognizing, telling apart, and writing of over 2000 glyphs. That's where the biggest "problem" and the source of most criticism seems to be - a lot of people aren't patient with doing so and feel like it's a waste.

TwoMoreCharacters - what's the point in spending 3 months not learning any actual Japanese though? I suppose the round about point I am making, which seems to agree with others here, is that to 'do RTK' whilst forsaking all other input (as AJATT seems to suggest is viable before then going on to mine sentences), cheats you out of getting the payback from the method.

I think there's a lot of point scoring on this forum, and elsewhere in the Japanese language learning world, where people want to say they did it for the sake of saying they did, rather than for the payback the method is sure to bring. I think the point isn't that people aren't patient, but that they don't make the method pay. They rote memorise RTK and then sit there wondering why they still can't read.

I think RTK is all well and good if you have a clear aim in mind, and don't get lost in the 'I must finish this' mentality. I stopped at about 1600 and have no intention of ever going back to finish it - I don't need to, I took from it what I needed and moved on.

I think we agree though in essence!

i don't get this 'spending all that time learning rtk and not able to read' nonsense. Anybody who is going into RTK 1 thinking they will exit the book with the ability to read japanese is an idiot that failed to even read the introduction. And let's be honest here, Heisig intended for this to be a fairly quick program to go through, stating about 3months as the middle of the road estimate for someone doing the book and having a working life. Myself, as I have no real obligations outside of the home (long story and not important) took me right at 1 month to get through the entirety of volume 1. And i was lazy for a good week of that time lol. Now that I finished volume 1, i'm beginning to add volume 3 kanji and keywords in, while going through Genki I. Yes, RTK might not make you able to know the correct pronunciations of kanji, but you will know a pretty good approximation of it's meaning, and i've found that quite helpful already. Even watching anime or tv quiz shows, i'm realizing much more of what all is going on (in written forms anyhow). The satisfaction and gleeful feeling of picking that kind of stuff out has been a really good motivation for me. I actually have more trouble remember the words in kana form than kanji at this point Also, there's no way i would've survived it w/o SRS. Let me walk you through how i studied rtk. first, i'd have the book and a drawing program with my wacom tablet ready. read kanji, write kanji, repeat writing a few times. then I would go to the next, usually completing a chapter in one sitting (minus a few of those massive chapters ..gah). Then i'd load this site up and add the kanji from that chapter and go through a process of adding stories. export file (in both stories and pure flashcard only formats) to import into anki. I actually have 2 anki decks for rtk. 1 is with stories and the other is w/o and i do both in one sitting. usually do the one on this site in the mornings, and those 2 in the evenings. It may sound insane, but it definitely worked for me so far. When i'm out of the house waiting at the bus stop, i always take my sketchbook (i'm an artist) and a small notebook, so I can think of keywords and try recall that way. people may look at it weird but hey, i'm learning.

@jankensan and others likeminded -

Doing RTK while forsaking other input doesn't 'cheat' you out of the payback, it postpones the payback until after you've completed it. The reason for forsaking other input is to complete RTK in the least possible time (as in hours of studying, not calendar time passing which is another matter altogether but also incidentally shortened).

The whole point of 'doing' RTK is to build up a set of fairly artificial associations with one clear meaning for each character and to embed a certain set of characters into your memory at one go. If you read other material and study words in their kanji form, you're muddying that clear meaning before all those associations are set and making your RTK work harder and take more time - more hours of study more failed reviews.

Of course, for myself, I'm not completely forsaking other input to -just- do RtK. I do enjoy the kanji for themselves, but... not that much to ignore all the rest! However, I am avoiding -reading- as much as possible, and instead watching dramas and listening to radio or podcasts, etc, more A/V and less print media input.

I also find nothing wrong at all with learning more kana vocabulary while doing RTK, though I haven't been particularly pursuing that, it does occur to me that pursuing onomatopoeia and katakana words is a great way to spend 'study' time that isn't RTK without hurting RTK progress, I may have to spend some time on that when I'm out of energy for Kanji stories!

All that said - if reading your manga or what have you is that important to you, go for it. It may mean more 'work' on both sides - as in less vocabulary retention and less RTK retention and more time studying those - but it's more important that it doesn't -feel- like work than the number of hours it takes, of course! That's exactly why I'm enjoying my video-watching and not being purely cloistered in with RTK.

(And I'm not really sharing any deeply secret stuff here, I thought this was all made pretty clear in the introduction, at least of the 4th edition that I have.)

Hmm... Rtk certainly isn't necessary in order to read/write Japanese. Trolling the Internet turned up the following.

Patrick Kenny who created the website Nihongo Pera Pera to help Japanese language learners, and works in Tokyo as a translator, stated "[Rtk] is popular on the Internet, but I've never heard of a Japanese teacher recommending Heisig." He wrote a review explaining how he started with Rtk but became disillusioned, turning to Kic instead.

Lloyd Vincent, another translator, living in Nagoya, who writes the blog Nihonshock simply learnt his kanji the traditional way, by rote. He explains why he preferred to do so here

Both these people have far better Japanese skills than me : (

I think it depends a little on one's personality. Heisig gave me an entry into the world of kanji and I think any other method would simply have been too daunting. I also appreciate that knowing the primitives, and a nemonic method to go with them, makes learning new kanji easy On the other hand, I sometimes worry about the amount of time I spend fretting over and reviewing similar English keywords.

I see it like this:

Let's say two people both have several bills to pay.

One person writes out a check.
Puts the check in the envelope.
Walks to the post office and sends it.

Comes back, writes out another check
Puts the check in the envelope.
Walks to the post office and sends it.

Etc.

This is the one-kanji-at-a-time Rote memorization method.

The other person writes out all of their checks.
Puts all the checks in envelopes.
Walks to the post office and sends them all.

This is the RTK method.
RTK is about writing out all the checks ahead of time.

Later you can worry about the envelopes.


But, they end up in the same situation in the end -- All their bills paid.

Manske Wrote:I did not use RTK. I think it is a waste of time. I straight up used flashcards to memorize on and kun readings for my first 1000 kanji, and after realizing how tedious and inefficient that method was I decided to just learn new kanji along with new words.

So how do you learn to write characters?

Anyone can learn to read recognition only - RtK is primarily a quick method of learning to write (with the added bonus of knowing simple meanings and being able to differentiate very similar kanji). If you just want to READ, you don't need it..

jankensan Wrote:TwoMoreCharacters - what's the point in spending 3 months not learning any actual Japanese though? I suppose the round about point I am making, which seems to agree with others here, is that to 'do RTK' whilst forsaking all other input (as AJATT seems to suggest is viable before then going on to mine sentences), cheats you out of getting the payback from the method.

I still don't get where this "do RtK and nothing else" meme came from? This seems to be one of the most common criticisms; why spend 3 months stories, when you could be learning actual Japanese? But it's not true - you can do Heisig and regular study at the same time..

I personally studied RtK alongside doing Japanese grammar and vocab in kana only (using Japanese for Busy People: Kana Edition). Then when finished it was just a matter of putting the pieces together - easy really.

Harpagornes Wrote:Lloyd Vincent, another translator, living in Nagoya, who writes the blog Nihonshock simply learnt his kanji the traditional way, by rote. He explains why he preferred to do so here

Copying out an entire dictionary multiple times is my idea of HELL! Interesting that it worked, but sounds like a tedious nighmare. Once thing haters seem to miss about RtK - it's a really fun method! I find myself LOL-ing daily at my stories, or other stories people have come up with. It's occasionally a grind reviewing, but overall I love it. Wonder if he could say the same the "dictionary" method?

When considering the efficiency of RTK, the most important question in my mind is are mnemonics necessary? Do they speed up the process?

The act of creating mnemonics undeniably takes time. Maybe it takes 10 seconds, maybe it takes a minute, maybe it takes 2. But obviously creating a mnemonic takes more time than not creating one. Multiply this time by 2000 and that would be 16.6 hours (based on a very low estimate of 30 seconds per card). So it would take about 17 days at an hour a day, not including review time and that would be if you're really fast and focused. Is this very conservative outlay of 17 hours necessary to remember how to read and write kanji compared to just reviewing them in a sensible manner? I'm tempted to say no. Are English keywords useful as anything other than a prompt for writing? Sometimes, but that same time could have been spent learning actual words or readings which are arguably much more useful.
It is easy to fall into the trap of thinking mnemonics are necessary if you insist on short interval reviewing of everything. When you mistakenly believe that it is vitally important to have 90%+ retention after one day than spending time crafting mnemonics appears to make a lot of sense. But if you abandon this type of short interval review, you'll realise that items start entering your long term memory quite effortlessly as long as you reencounter them in the longterm. So it requires much less time per item to have an organised system of long term review. The trade off is that instead of being absolutely sure you'll know the item after 1 day, you can be absolutely sure that you will know it after 1 month. So you have to wait a little longer in the short term but spend significantly less time on a per item basis, meaning you can cover more material given the same amount of study time.

I am studying Mandarin now and in the period of May 19th to July 24 of this year (about 2 months), I have studied 2700+ words and am picking up the writing and readings for free as well as getting a sense for the meanings associated with individual characters. My recall in the long run is no worse than if I had been using mnemonics.

People seem to regard RTK as the only efficient way to literacy and compare it to traditional classroom methods. The problem with classroom methods isn't necessary the lack of mnemonics, but rather the total amount of kanji covered, rate of introduction, encouragement of redundant short term review (re-writing again and again) and ill considered longterm review (quizzes and exams).

Obviously Mnemonics are not the only way to do it but when you're talking about someone with no Kanji knowledge who wants to approach Kanji, I argue that it's one of the most effective.

You mentioned learning 2700+ words but I bet you already knew a lot of the Hanzi and you had a ton of the fundemental Kanji base from (Assumedly) doing RTK or whatever other method you used to learn Japanese Kanji. I don't think Mnemonics are best for vocab, but Kanji are invariably different than vocab.

zachandhobbes Wrote:Obviously Mnemonics are not the only way to do it but when you're talking about someone with no Kanji knowledge who wants to approach Kanji, I argue that it's one of the most effective.

You mentioned learning 2700+ words but I bet you already knew a lot of the Hanzi and you had a ton of the fundemental Kanji base from (Assumedly) doing RTK or whatever other method you used to learn Japanese Kanji. I don't think Mnemonics are best for vocab, but Kanji are invariably different than vocab.

Kanji is actually much easier that some seem to want to make it seem. Depending on the definition used Kanji is mnemonics by nature. RTK is a good method for learning Kanji, but some seem to forget that they are learning Japanese.

My general approach is to try to use vocabulary known to learn new vocabulary, which does include using mnemonics. However, if it is to learning one or more English equivalence for Japanese words then it could be come difficult.

From my perception RTK's mnemonics method removes the intimidation factor in learning Kanji, by allowing students to use what they already know. There are various methods to learning Kanji, but outside Japan it seems that RTK is the most famous.

Harpagornes Wrote:Hmm... Rtk certainly isn't necessary in order to read/write Japanese. Trolling the Internet turned up the following.

It is true. Learning Japanese is possible without being able to write the characters per hand. The IME does wonders and writing is possible. But the distinction between similar Kanji after having gone through the process, sharpening the eye for the for those small differences that otherwise might lead to confusion, shows that it is a good idea to learn the RTK way. Or any way at all in an efficient and timely manner.

Harpagornes Wrote:Both these people have far better Japanese skills than me : (

Dont ever start to measure anyones progress to your own. Not everyone has the time to get there in 2 or 3 years. Well, some geniuses can, as for the rest of us we have to deal with a lack of time, other commitments, work, school, you name it. It is a commitment to learn a language, and there are people out there for whom it took 10 years, others were able to get there in 1 or 2 years. Both can succeed or fail. Had they compared themselves to the other guy, they might as well have quit, because it seemingly takes forever to get were they are. My best advice to you is, don't start to compare your progress to anyone elses, but look at what you were able to achieve up to this point.

Quote:I think it depends a little on one's personality. Heisig gave me an entry into the world of kanji and I think any other method would simply have been too daunting. I also appreciate that knowing the primitives, and a nemonic method to go with them, makes learning new kanji easy On the other hand, I sometimes worry about the amount of time I spend fretting over and reviewing similar English keywords.

The keywords are no harm at all. They are supposed to fall away, and they will, so don't worry about it.

aphasiac Wrote:I still don't get where this "do RtK and nothing else" meme came from? This seems to be one of the most common criticisms; why spend 3 months stories, when you could be learning actual Japanese? But it's not true - you can do Heisig and regular study at the same time..

When only doing RTK there is more time to learn a larger quantity in one sitting. And 3 months or less doing RTK does not hurt. I don't think that anyone should take this advice too serious and not learn anything on the side.

aphasiac Wrote:I personally studied RtK alongside doing Japanese grammar and vocab in kana only (using Japanese for Busy People: Kana Edition). Then when finished it was just a matter of putting the pieces together - easy really.

I did Genki before moving on to RTK. And RTK was all i did besides listening and watching Anime with subtitles for almost 3 months. Don't forget that not everyone learning the language has the mental capacity, the iron will, or time to do both at a time.

I am completely sold on the Heiseg method. I tried learning both RTK and grammar alongside each other for a few months but having to write everything in Kana or half and half just seemed stupid(also took a while).

With my limited hours per day to study I just decided to concentrate on completing RTK as soon as possible. Now that I'm done I can definitely say it paid off. Seeing how kanji come together to make up their words is pretty cool. E.g. 宿題 at first glance I see that as 'inn topic' when it is actually homework.

Frankly I don't see why people would quit after a few hundred, one of the biggest pushes for me was that if I'd made it this far why stop (throwing away all that effort). Seems to me people are looking for a drastic shortcut when there isn't one. I can certainly say I enjoyed the journey then again I did the entire of RTK listening to music.

Kysen Wrote:I am completely sold on the Heiseg method. I tried learning both RTK and grammar alongside each other for a few months but having to write everything in Kana or half and half just seemed stupid(also took a while).

That just shows that splitting your time is inefficient, not that RTK is the fastest way to learn kanji. There are other ways of systematically learning 2000 odd kanji than the methods in RTK.

Kysen Wrote:With my limited hours per day to study I just decided to concentrate on completing RTK as soon as possible.

Why not concentrate on learning 2000 odd jouyou kanji the fastest way possible? Ie. focus on the material you need to learn and evaluate if the RTK method really is the fastest way to do that.

When you really think about it, RTK really is akin to learning the spelling of 2000 odd english words without learning how to pronounce them...

I don't think even Heisig realized it at the time, but for native learners or second language learners, RTK (RevTK, rather) is pretty much the best way to learn kanji. It has everything that's ideal for long-term retention and which minimizes the time you achieve that superior long-term retention: the mnemonics, which are better for long-term learning in less time than slow, shallow learning that takes longer and seems nice in the short-term but is terrible for the long-term, it combines these transient relational structures with retrieval where you have weak cues that are perfect for keyword→kanji, you've got the spacing effect, and you've got the detailed, robust, bottom-up visualization of radicals/primitives to the instantaneous viewing of the kanji as a gestalt whole.

Combine it with the assembly line approach that minimizes overhead and takes advantage of the transfer-appropriate complementary nature of the kanji as meaning-oriented icons and contextualized meaningful readings that you can mix and match how you like, and you've got the best method for learning kanji for any type of learner, and one that gives you the skills to learn new kanji effectively and robustly even after you've moved on to only occasionally learning new kanji in context.

nadiatims Wrote:When you really think about it, RTK really is akin to learning the spelling of 2000 odd english words without learning how to pronounce them...

I don't think comparing old english to ideograms is necessarily an equivalence you can make. maybe ancient egyptian. :\ and that's the key problem here, for the vast majority of people using RTK, they are coming into it from a language that does not function in the format of ideograms. That takes a lot of mind warping to get comfortable with (and hence the biggest purpose of RTK besides writing, comfort witih the concept of ideograms).

Yes, your Japanese is much better than mine, but Japanese have been learning Kanji long before Heisig became a student of that language. Heisig was a student of Japanese when he first thought about his mnemonics for learning Kanji. Yes, mnemonics are very helpful for assisting both native speakers, and non-native speakers learn a particular language. Such as 「大きな羊が美しい」 in Japanese for 美. Also, by its very nature that is what Japanese is all about. Take for example 「人の動くは働く」. 働く is just one of many 国字. Many rules for Kanji are beyond the intended scope of RTK. However, that doesn't diminish the fact that RTK is a good method to remember Kanji.

nadiatims Wrote:I am studying Mandarin now and in the period of May 19th to July 24 of this year (about 2 months), I have studied 2700+ words and am picking up the writing and readings for free as well as getting a sense for the meanings associated with individual characters. My recall in the long run is no worse than if I had been using mnemonics.

How do you learn the writing without Hesig? Can you detail your exact method? Cheers!

nadiatims Wrote:People seem to regard RTK as the only efficient way to literacy and compare it to traditional classroom methods. The problem with classroom methods isn't necessary the lack of mnemonics, but rather the total amount of kanji covered, rate of introduction, encouragement of redundant short term review (re-writing again and again) and ill considered longterm review (quizzes and exams).

Indeed - mnemonics are only part of the Hesig method. Certainly they're nothing new (my Taiwanese girlfriend said she used them for learning the harder hanzi), and also not necessarily needed either (there's been a number of threads on here for people who have finished RtK without using mnemonics).

I'd argue it's more about the ordering, breaking down of primitives, and use of unique keywords to fix each character in your brain.

Maybe I missed it Nadiatms, but you seem very critical about RTK not being a great method. I find it hard to believe that RTK isn't good seeing my own results, so I'd inqure to know exactly what your method is...

aphasiac Wrote:How do you learn the writing without Hesig? Can you detail your exact method? Cheers!

I already know how to write most components from having studied Japanese and when I encounter new components, I just pick them up by writing out the words as part of my vocabulary study. If I didn't know kanji, I'd still do the same thing but I might use Heisig's book to learn the stroke orders.

I've outlined my method I number of times on this forum. Here it is again:

1. Buy a note book (I use 30 lined (A4?) 40 double page notebooks from convenience stores)
2. Open to the next double page (first time, this will be the inside cover(left) and the first page(right))
3. Write down a 3 column list of words (up to about 30 items)you want to learn (first iteration, this will be on the left half of the left page.) 1st column is word in hanzi, 2nd is word in pinyin, 3rd is meaning in English or Japanese.
4. write down the date above the list.
5. ignore that list for a good chunk of time (I suggest at least 2-3 weeks)
6. Go back to the list and check which words you still remember, by covering the english and pinyin and reading the Hanzi.
7. Relist the words you didn't remember (again using the 3 columns). If this is the second iteration, you should use the right half of the write page. 3rd, 4rth and subsequent iteration will use any remaining space left on the double page. Don't worry there'll be plenty of space, because the list get shorter with each iteration.
8. Repeat steps 4 - 7 for the words on each double page.

Each double page of the notebook is a space for reiterating a list of up to about 30 items.
Keep reiterating the list until all the items are remembered.
Optional: When the lists get too short, less than say 5 words, you can start combining multiple very short lists onto a new double page.

Keep writing new word lists at a pace that is comfortable for you and keep reiterating the old lists until they are done. Most of the items will be learned by the 3rd or 4rth iteration, ie. after having written it up to 4 times extended over a period of 1~2 months. If each iteration of an item represents 30 seconds, then that means most items are learned in a total time of 30 ~ 120 seconds each.

You can calculate the number of items in each notebook by multiplying the number of double pages (40 + inside cover = 41) by the number of items in each list (30ish) amd write that on the front of the notebook to track how many items each book represents. In my case 1230ish items per notebook.

It's that just like using an SRS program, but manually, and with guessing the time intervals? So why not just use an SRS program?

EDIT: Also where does writing come into this method? Only seems you write in step 1, then you do recognition only? How do you test whether you can write the hanzi correctly from memory?

It's like using an SRS program but minus short intervals. First review is after at least 2~3 weeks. Words stop being reviewed after the first successful long interval review (and you don't have to delete cards in order to stop seeing them). The items are written at each iteration.

The process of writing words and characters is ingrained in time by the process of writing words. There is a huge amount of recursion because the same characters/primitives/radicals pop up again and again in different vocabulary.

How relevant is the ability to write characters from memory when prompted by an English keyword? This ability is not really useful to the learning process. When reading, the most important thing is correct recognition of characters and words. Dictionary look-ups can be done using pinyin or pop-up dictionaries or referring to translations (phrasebooks, parallel texts). If/When the character needs to be inputted manually you will generally have it in front of you in whatever source you found it in, so you can just copy it.