I am still not seeing what good offering the ability to change the keywords to Japanese is going to do outside of encouraging people who haven't finished the book to switch to Japanese keywords, thinking it is a good idea. They are far more important than those that have finished. If you don't like studying from English keyword to kanji, why not just add a flashcard in any one of the hundreds of formats available to your own deck to drill it however you want? Then you can do it synewave's way, Jarvick's way, yorkii's way, front to back, back to front, only on Tuesdays, whatever you want.

I don't know about you all, but when I read something below my level, like say: 大きな字で書く I read it as it is and understand it fine. I don't think anything about "large", "character", or whatever the heck the keyword for かく was (I think it was "write"). I also have no trouble writing any of those characters even though I haven't bothered to write them in a good long while. In fact, thanks to the way an SRS works, those cards are probably scheduled somewhere in like 2012 by now, problem of shedding the keyword solved.

At the same time when I read something like this: ある栄養士養成学校の学生を対象に行った調査によると、．．．(assuming I didn't know 栄養士) I am more likely to think (reading I don't know)"flourish",(よう)"foster",(し)"gentleman" (something about a person who fosters and or possibly helps someone to flourish) <the rest of the phrase>. In this case I think the English keywords have acted a safety net and consequently saved me a trip to my dictionary. I will highlight the word for later addition as a flash card where, if I were so inclined I could review from English or Japanese meaning to kanji compound and get writing practice in for three kanji at once.

I think the issue of changing keywords to Japanese takes care of itself the more you study after finishing RTK1, and as you reduce the time you devote to RTK1 drilling. I think an LCD Japanese keyword discussion, therefore, really just becomes an academic exercise for those capable of producing such a list. Sharing vocab is great, sharing methods to shed English keywords after you have moved beyond RTK1 is fine, but what good can come of trying to find a LCD Japanese keyword solution, even if the deck is in a separate area of the site, when this very thread has wonderfully displayed how varied people's needs become after RTK1? Why not just focus on the "after" in "after RTK1"? Isn't that enough?

dilandau23 Wrote:I also have no trouble writing any of those characters even though I haven't bothered to write them in a good long while.

This is the crucial point I think. I'm in London now so I don't do it that much these days but when I was in Japan, I spent a lot of time writing Japanese (on paper). After reading this sentence, it appears that you don't. I invite you to try and see what happens.

What I found was that despite doing very well reviewing from English keyword to Japanese, I'd encounter problems when writing sentences. When I was in the middle of a paragraph and wanted to write a simple word like 最後, I'd know which characters I wanted to write but I had to think carefully about their shape. This is exactly the situation that Heisig's method should help you with. Instead, I'd have to think about "What's the English keyword for this character?" "What's the story associated with this English keyword?". Despite knowing the character, my recall of the proper English keyword sometimes took a long time. What's worse, many of my stories and many of Heisig's stories use meanings of English keywords not associated with the kanji. eg. The baseball team and the car, stuck in a rut etc.

Now that I've been reviewing using Japanese keywords, I'm used to going straight from the hiragana to the kanji via the imagery from my story. I no longer have these mental mini-mountains to get over every time I'm writing a word I'm unsure about. Instead of

hiragana -> English keyword -> English story -> kanji
it's
hiragana -> story imagery -> kanji

I hope that this helps explain why I think the English keywords should be scrapped. There are lots of minor complaints that can be made but I think that this is THE major one. When you're writing in Japanese, they get in the way.

I invite all RTK graduates who can write in Japanese to try it. The next time you send a Japanese email or whatever, write it down in Japanese first and see what happens. If you all are fine and don't encounter problems then fair enough but personally, my life has become a lot easier after reviewing my Heisig stories using Japanese keywords as prompts.

Personally, I think there are much better things that could be done first before we add Japanese keywords.

-For each primitive, show how many times it appears again in RtK 1 (maybe RtK 3 also, but I know Heisig goes back and redoes primitives so that may be difficult). If there are different versions of the primitive (大), display how many times each version appears. Then, allow users to search for primitive, so if they are struggling with a primitive they can focus on it, or even maybe try to create a chain of stories, now seeing several kanji using the primitive. (I have some weird image calling the right primitive on 模 "The sun dog" and continue to use this for every kanji it appears in. Kanji using this primitive I am usually able to recall 100%, and I'm sure it's because of the chain of memorable stories.)

-Similar to the practice page creator, allow users to export .doc file containing a simple list of keywords or kanji. Allow the option to designate the list by numbers or by exporting failed kanji.


Anyway, aside from my personal wishes, as far as Japanese keywords go and new users being tempted to do them, just do not make them visible until you already complete RtK1. Do not let users know (explicitly) of the option until they complete RtK 1. I don't think many will try to search the option out or switch to it because they think it will be better. In fact, I think many people who wouldn't be ready for Japanese keywords would just instead be intimidated and opt for the English words given the option.

sutebun Wrote:Anyway, aside from my personal wishes, as far as Japanese keywords go and new users being tempted to do them, just do not make them visible until you already complete RtK1. Do not let users know (explicitly) of the option until they complete RtK 1. I don't think many will try to search the option out or switch to it because they think it will be better. In fact, I think many people who wouldn't be ready for Japanese keywords would just instead be intimidated and opt for the English words given the option.

Don't have the time right now to respond to everything since my last post, but I'll do this one. On the surface what you wrote seems like a good idea. But what about people who have studied Japanese before coming to RTK? I think there are a good portion of people who have vocabularies much stronger than their kanji skill. That kind of person could and should start with Japanese keywords directly.

[ファブリス]

wrightak Wrote:I cannot understand why you think 山をのぼる would be so much more of an information overload than のぼる.

Fair enough, but I'm tryintg to think ahead. This is the shortest example possible with only two characters added to provide sufficient context for the compound. Anyway I think we're agreeing towards synewave's suggestion, very short prompts where the word alone is not sufficient.


So far, it looks like most of the arguments for Japanese keywords come from people who have finished RtK, and who have a good basic level in Japanese.

With many frequent words having entered your visual memory, it's logical that you may think of a Japanese word and remember the associated character(s) without an intermediary step through the english keyword.

I think that's the perceived benefit of using the Japanese keyword : the keyword (or short prompt) provides additional visual memory links, where the english keyword doesn't.

So as such, it's true that this step is not really necessary if you move onto reviewing plenty of vocabulary. But in practice, it takes a while to be exposed to the variety of kanji in RtK1 and absorb all the extra vocabulary and grammar around it, so I think the Japanese "keyword" review could be really helpful.

I'll have to think more about how to handle the transition for someone completing RtK1, to the Japanese "keyword" list.

Are we agreeing on the kunyomi readings though?

> I think it would be best to have only kunyomi if possible, because those are the ones for which RtK2 doesn't provide a really practical solution. Onyomi can be handled by a variety of programs, and is also covered in RtK2.

> Another point is the user might be tempted in guessing the character from the compound if they know they are onyomi, since onyomi is often related to the components of the character; whereas kunyomi is not so organized.

I am a relative newcomer to RTK, but not to the Japanese language. Jarvik7 made the point that many people come to RTK with a fair amount of spoken Japanese or having studied the language before. I am truly puzzled by people's concern that hapless newcomers will lose their way by switching to Japanese keywords too soon. For the most part we are talking about cognitively normal adults, aren't we? And these adults can assess their own situation and make decisions for themselves (with the expert guidance provided by this site), I would think. I suspect there are a substantial number of people who find that there is an extra mental step (as Wrightak has eloquently pointed out) in working from English keywords. The goal for me is not finishing RTK1 (although I am working slowly, but steadily, at this), but proficient reading ability.

After sleeping on it last night, I started to wonder if it would be possible to limit Japanese keywords to 1 or 2 words all the time. And if this isn't possible is it preferable to have a longer Japanese prompt (key sentence) or use some English?

EDIT: not only that, but if we are going to use Japanese (which I think is the ideal), users may encounter issues, not because they don't know a keyword but because the additional Japanese context is unknown vocabulary. So now I'm beginning to see the merit of English clarification...[/EDIT]

Can anyone suggest a good prompt for 且つ？

The best I have come up with so far is 「かつ（接続詞）」。Is it too long? Too "complex"?

What level of Japanese knowledge is to be assumed for users going over to Japanese keywords?

ファブリス Wrote:Are we agreeing on the kunyomi readings though?

Yes. As far as possible.

synewave Wrote:EDIT: not only that, but if we are going to use Japanese (which I think is the ideal), users may encounter issues, not because they don't know a keyword but because the additional Japanese context is unknown vocabulary. So now I'm beginning to see the merit of English clarification...[/EDIT]

I see what you're getting at, I just think that if someone needs English clarifications for Japanese keywords, then they aren't ready for Japanese keywords. Their vocab just isn't wide enough to make it practical. I would find them useful too, and I know I'm not anywhere near ready to make the switch.

Keep Japanese keywords geared towards advanced students, leave the rest of us wrangling with English keywords.

dilandau23 Wrote:I think the issue of changing keywords to Japanese takes care of itself the more you study after finishing RTK1, and as you reduce the time you devote to RTK1 drilling.

This is probably true for a lot of characters but not necessarily all. So the idea of systematically being able to study from 訓読み to kanji is appealing.

dilandau23 Wrote:I think an LCD Japanese keyword discussion, therefore, really just becomes an academic exercise for those capable of producing such a list.

I don't know what LCD means in this context :|

synewave Wrote:

yorkii Wrote:I'm not even talking about sentences really, just sufficient context for one to know which kanji you need to write when there is ambiguity. that's all.

So if there is no ambiguity, e.g. こころよい, as far as I am aware, am I right in thinking that you would be happy enough with this single word as the Japanese keyword to prompt you to write 快（い）？

yes, that would be the ideal

Wrightak, you bring up some good points that I wasn't considering. In particular, the bit about not being in Japan. I suppose I can very easily see the merit of Japanese keywords if I were to place the Japanese student outside of Japan where they may not have the time to commit to studying that they once did.

Also, to be honest, learning at least one word for every RTK kanji does seem like something I could imagine a student wanting to do after RTK1. It's systematic, it could easily seem comfortable to the student. It can help the advanced students no longer exposed to a constant flow of Japanese. I just feel that at the point when someone is ready to digest such a list, it would be more productive to learn, or first time review, them in an order different from RTK1 and also not place any restrictions on the words they decide to learn. Namely, put all the things like 且つ at the end and disregard concern for homonyms(in the context of choosing words not in regards to being aware of their existence). I think synewave and Jarvik have the most elegant solutions possible for that anyway (I lean towards keeping the cues in Japanese thought). As for 人名用漢字 I think that you may as well put a name if that is where the kanji is used most. Admittedly, these choices are just personal and may not suit anyone else.

I can tell you that what I do post RTK1 to work on my writing. While I haven't been doing any of this lately because of the looming JLPT, most of my flashcards are created from things I read that I struggle with in real life, say in a book, or from the forest of paperwork that gets put on my desk. At the same time I try to pull about 5-10 compounds(I know some of them are pretty obscure so I pick my own if I feel the need) from RTK2 and add those to an RTK2 model that I made for Anki. When I see a compound I write it down whenever I don't remember its reading within a reasonable amount of time. This really helps me to merge the compound to the reading and writing it also helps me to remember the compound a little easier the next time around. Outside of that, I almost never handwrite anything but my address. I don't have a need to, so I don't worry about it much.

I will close with this last bit and then I promise not to try and rain on your project anymore. Yes, like Laura says, we are cognitive normal adults. I just know that before I did RTK1 the way it was designed, I was a cognitive normal adult and I had no idea how to tackle kanji. I tried and failed several times. If I had come here and seen all these people much farther along in there studies complaining about English keywords "getting in the way" and making a switch to Japanese ones, I may have been tempted to cut out the middle man. I can't say what the end result of that may have been, but my instinct tells me that I would have failed at it as well. I may have even just written off RTK altogether. Once you are far enough from your starting shore, it is easy to forget what it was like. I just implore you all to not forget about the new student.

EDIT to answer syenwave: Lowest common denominator

Are these the sort of keywords that people envisage?

1 one ひとつ
2 two ふたつ
3 three みつ
4 four よっつ
5 five いつつ
6 six むっつ
7 seven ななつ
8 eight やっつ
9 nine ここのつ
10 ten とお
11 mouth くち
12 day ひ or ひにち
13 month つき
14 rice field た or たんぼ
15 eye め or めをとじて

yes.

おつきさま would probably be better in this case because there is of course 付き etc.

It will be interesting to see wrightak's list. Does it contain a mix of 訓 and 音読み？ How sentence based is it?

What should we do if no good "short" phrase can be found but clarification is necessary?

What should be done for characters with no 訓読み？ What about rare 訓読み？

Initially I though this would be fairly easy but the more I look into it, it seems that it will be fairly time consuming...

It would be nice if we had a wiki of some sort. We could all collaborate and get RTK1+3 wordlists done pretty quickly.

i suggested a "google document" which we would all be able to collaberate on earlier. they have spreadsheets and things that would be very easy to implement. I could be wrong here, but I think it was wrightak that had already opened one for a different project.

To me, the most important thing is to have an unambiguous keyword, regardless of which language it is. For numbers, "six" or "6" is better than 「ろく」. There's nothing ambiguous about "6", whereas I wouldn't be sure which 「ろく」 I'm supposed to use. A lot of the English keywords are fine, especially the nouns, but sometimes I can't remember which kanji the keyword represents. I know the kanji, what is means, how its pronounced, etc., but I've lost the link between the English keyword and the kanji. I don't really care about that (in fact it's a good thing), but I can't review the writing of those kanji properly because I have to see the kanji to know which one I am supposed to write. Having both English and Japanese to look at can help to reduce ambiguity, and I wouldn't think there would be much harm in just having both appear on the cards. Another way to handle it might be to have an option for each card to use the English keyword, Japanese keyword, or both.

I think a Wiki sort of thing is fine, although I wouldn't wait for every last kanji to have a Japanese equivalent before you implement it. If you put out a few hundred cards with Japanese, it would encourage people to help finish the rest.

I think google documents would be better for several reasons. Firstly, several people can collaborate on the spreadsheet at the same time and they can see the changes that other users are making as and when they make them. Each user is assigned a colour and the cell they have selected is highlighted with that colour. Secondly, you can chat with the other collaborators to discuss things when you're editing the document at the same time.

Synewave is right, it's a surprisingly difficult task but I'm hoping that with many people collaborating, it will be finished fairly quickly. I haven't had time to get it all together yet and I'm sorry about that but I reckon I'll have it done by tomorrow lunch time UK time.

brose Wrote:There's nothing ambiguous about "6", whereas I wouldn't be sure which 「ろく」 I'm supposed to use.

Out of interest, what other ろく do you have in mind? 録？

[ファブリス]

brose Wrote:Having both English and Japanese to look at can help to reduce ambiguity, and I wouldn't think there would be much harm in just having both appear on the cards. Another way to handle it might be to have an option for each card to use the English keyword, Japanese keyword, or both.

Having both on the flashcards is reasonable but there are the other problems..

I see this most fruitful as a kunyomi review for post-RtK1 learners, to keep on reviewing characters AND learn 2000 words at the same time. Hit two birds with one stone.

In this context, you wouldn't want to see the english keyword since you also want to actively recall the meaning from the Japanese phonetic reading. It's one of the review options mentionned by Heisig at the end of RtK Vol 2.

OK, sorry for the delay, I finally got around to organising it all. I thought that I had two lists in addition to my own but I only found one from raulir. You can see it here:

http://spreadsheets.google.com/pub?key=p...S9s2aQ5xRQ

I tried to be clever and use a vlookup function to merge it into my spreadsheet but for some reason that I am unaware of, the vlookup function refused to work properly on the data. If someone else can figure out how to do it then please feel free.

In addition, I should add raulir's disclaimer that he sent to me:

Raulir Wrote:I pretty much threw away my guidelines in the middle of making it and
started using anything that works for me, so it may not be all that
useful for your purposes. No harm in having it, though :-) Some of the
examples and kanjis aren't, uh, "family friendly", as that has never
been a goal of mine, so don't be surprised if you find a couple of
those.

Below, you can have a look at my spreadsheet. I've put the words on one sheet and example sentences on another sheet. If people would like to collaborate with me, or if you'd like to be able to view it in excel format then just send me an email and I'll be happy to send you an invitation.

http://spreadsheets.google.com/pub?key=p...UHNQ&gid=0

There's space in the spreadsheet for three different words for each kanji. So if people think that another word would be more appropriate, you can put the suggestion in the Word 2 column or the Word 3 column. I also added three columns so that people could vote for which one they think is best.

The one thing that I would ask of anyone who wants to collaborate on this is to please put the date of any changes you make into the "Last Updated" column and put your name beside that. That way, we can keep track of what's new and what's old.

ファブリス Wrote:I see this most fruitful as a kunyomi review for post-RtK1 learners, to keep on reviewing characters AND learn 2000 words at the same time. Hit two birds with one stone.

In this context, you wouldn't want to see the english keyword since you also want to actively recall the meaning from the Japanese phonetic reading. It's one of the review options mentionned by Heisig at the end of RtK Vol 2.

I am not interested in trying to learn English equivalents of Japanese words while reviewing the kanji. This is especially a bad idea for the Heisig keywords. The idea for me is to use Japanese words I already know, and see if I can write the kanji. That's also why I think the common reading なつ（かしい）should be used for　懐. I don't need to know how to write words I won't use.

Wrightak, Thank you! The Japanese keywords are like getting a present. I tried coming up with my own list and got sort of boggled. And being from the generation that considered the IBM Selectric "high tech," trying to invent my own stacks, or program, would take up all the mental energy I have for learning kanji.

wrightak Wrote:If people would like to collaborate with me, or if you'd like to be able to view it in excel format then just send me an email and I'll be happy to send you an invitation.

Done.

ファブリス Wrote:I see this most fruitful as a kunyomi review for post-RtK1 learners, to keep on reviewing characters AND learn 2000 words at the same time. Hit two birds with one stone.

If we wanna see this become part of the site, we better do it like the man says. I'm all for the idea anyway

I just looked at both lists, and I think that they are a great start. One thing bothers me about them though, which will probably change as we all refine them further and further. I think there is way too much extraneous information on the charts as they are now, it's very distracting. I realize that as we work with these, they'll probably become more streamlined, but I really hope the finished product will allow us to focus on bonding just ONE word/short phrase with ONE kanji (or, if it REALLY is the case that a 熟語 is a good choice to remember an individual kanji by, then a compound). Too much extraneous information such as lists of 音読み will become very distracting, and make it very hard to make the images and stories that make this method work. I hope that the final form of this project will be to have 2042 (hopefully soon 3007) 'pages', each with a kanji and a japanese word on it, for us to in a way redo RTK1, in Japanese this time.

I know that people probably would have worked out what Im talking about anyway, but I felt it was important enough to mention.