Wrightak, a quick question: a lot of your sentences have 亘 used as a word (for example, 240, 295, 420, 979, 1093, 1412, 1721, 1836), but I can't find it in Sanseido nor in EDICT. The verb form わたる is of course there, but the kanji alone isn't... are you sure it's a common word?

Ha! I should have thought of that! Thanks.

I finished the book only 4 months ago after reading it for more than a year almost exclusively on the morning commute, overlapping in great part with my study of RTK1. The reason I remember the names that well is that I struggled a good deal to remember them. They're only ever written once in furigana when a character is first introduced. Forgetting one and having to locate it again in a book that's over 1000 pages (divided into 3 pocket books) is a nasty task. A dictionary is only half a solution as the same name in kanji can be read in so many ways. E.g. is 香織 かおる or かおり?

I thought I remembered the title mentioned here and searching the forums I found a message from you dating one and a half year back, around the time I bought the first part. That could have been my trigger. (Seems you took your time as well: another message dated 9 months later still mentions "the book I'm reading".)

I haven't seen the movie yet, but it looks attractive to buy, now that I've finished it.

Last edited by woelpad (2008 July 08, 2:24 am)

Wrightak-- did you use the Tanuki corpus at all for the Japanese keywords? I had completely forgotten I still had a copy of that lying around on my desktop. That would probably save some time in setting up my deck/spreadsheet.

Yeah, I was trying (in vain) to find anything else useful for 亘 besides 亘る, but even the Tanuki doesn't cover that kanji, which surprises me. It seems to cover a ton of kanji. I think I may plunder it for my RTK2 JP keywords thing, anyway. My only concern is the number of obscure/useless/weird kanji that are in RTK1 and 2 that will wind up wasting precious time. (Like 亘.)

rich_f wrote:

... 亘る, but even the Tanuki doesn't cover that kanji,

The Tanuki database covers the 1,945 Joyo kanji only (as main entries) but 亘 is a name kanji.

rich_f wrote:

I think I may plunder it for my RTK2 JP keywords thing, anyway. My only concern is the number of obscure/useless/weird kanji that are in RTK1 and 2 that will wind up wasting precious time. (Like 亘.)

The Tanuki database is in approximate order of usefulness. The order is taken from the "Kanji in Context" books (pub: Japan Times).

Levels as follows (I've simplified the descriptions and modified the numbers slightly to fit Tanuki):

1-249 Beginner
250-349 pre-intermediate
350-1,199 intermediate
1,200-1,418 intermediate to advanced
1.419-1,830 advanced
1,831-1,945 specialized fields

Yikes. Haven't had much time to experiment. I've been so busy with work that I'm just keeping up with my current reviews. But I am going to start carving out an hour or two when I can to add new RTK2 stuff.

I'm trying to set up a pattern to look for replacement words when Heisig gives out weird ones. So far, my search pattern involves:

0. Check the JP Keywords sheet to avoid reinventing wheel.
1. Spreadsheet containing all JLPT 2-4級 words
2. EPWING Dictionaries
3. Dictionary websites
4. Handheld dictionary
5. Nelson's Unwieldy and Completely Inefficient Book of Kanji
6. Give up and use Heisig's RTK2 vocab.

Since I'm trying to capture the pure/semi-pure/mixed flavor of RTK2, I can't use a lot of the Japanese Keywords... which is a little frustrating at this point, but it's early in the experiment. Also, I'm finding that some kanji with some readings are just insanely obscure or rare, so to save my sanity, I just use Heisig's examples. They're not all THAT bad, anyway.

But using 沖 for ちゅう when 99% of the time it's おき is a bit nuts... especially when it's in the first "pure" group. There's like 2-3 words for the ちゅう reading, and about a hundred or more for the おき reading... I get what he's doing, but I also am left scratching my head a bit.

Here's an anki plugin you can use to populate an extra Japanese prompt field with data from another file, such as the Japanese Keywords spreadsheet or a JLPT vocab list etc: http://ichi2.net/anki/wiki/Plugins#overwrite-fields

wrightak wrote:

Could you upload both this list and the JLPT vocab list somewhere? I'd like to take a look.

Do the 4, 3, 2 and 1 refer to 4級、3級、2級 and 1級？

the vocab lists are from http://www.thbz.org/kanjimots/jlpt.php3

here are my files:
http://www.uploading.com/files/WOLY3VDD … d.txt.html
http://www.uploading.com/files/Z3UDQYDQ … t.txt.html

merged has 3 tab-separated fields: heisig number, reading and word.  the reading and word fields are sorted by level and have the level number before the first word for that level

split has alternating reading and word fields for each level to make it easier to ignore the ones you aren't interested in

the nice thing about using this for japanese prompts is that you don't have to go around looking for suitable words for each kanji like we've been doing, disambiguation through multiple words, multiple readings, and you can see what words you are supposed to know for your jlpt level

also, I was thinking of formatting the question fields like this instead:

4 ◯ぶ/あそぶ; 3 ◯び/あそび; 2 ◯園地/ゆうえんち; 1 ◯牧/ゆうぼく

To me, it seems like once you get yourself enough into the language, you're going to want to stop reviewing the kanji from a purely kanji standpoint anyway and just maintain the associations you've built with them in the language. With that, I can't see the purpose in wasting a bunch of extra time going through everything again and assigning new keywords.

But I've always been a minimalist when it comes to seemingly unnecessary effort.

Japanese keywords makes sense to me... but I understand the recommendation is to go through RTK1 all over again with Japanese keywords... I don't know if I can stomach the thought of going through the whole thing again. Is it really worth it?

I can really sympathise with how you feel about not wanting to review the English keywords anymore. I don't think you want to waste your time reviewing decameron -> 旬 etc. any more than you have to. You want to continue practice the writing but not with English words that are becoming a hindrance rather than a help.

This project is just about helping you to do that in the most efficient way possible. The largest hurdle for RTK 1 is inventing vivid stories and images for each kanji. You don't have to do that again. All you have to do is learn a Japanese word, delete the old English keyword card and replace it with the new word. You're taking away a card for every one you add so the only extra burden is learning the new word.

You can do it in a haphazard way, as Jarvik suggested, or by using words that you encounter, but I found that the most efficient method for me was to go through in Heisig order again. See what works for you.

It's not really about being worth it or not, because eventually everyone wants to use the work they put into RTK 1 in real Japanese. Let's put reading to one side and think about writing. What value is it to be able to write '申' when a flashcard shows you 'speaketh'? It's OK as an intermediate stage but you want to be able to write the kanji when you want to say もうします (申します). So gaining the ability is definitely worth it, whatever method you choose.

SammyB wrote:

So I've finished RTK1 now and have been getting stuck into sentences. I've noticed that when reviewing the English keywords now, sometimes I wont remember the kanji, but when I fail the card and it shows up, I can think of a Japanese word that I associate it with the kanji... It's kinda frustrating having to reinforce the English keyword association.

Japanese keywords makes sense to me... but I understand the recommendation is to go through RTK1 all over again with Japanese keywords... I don't know if I can stomach the thought of going through the whole thing again. Is it really worth it?

What I've been doing is replacing the English keyword with a Japanese one in my Anki deck as they come up for review. I do this only for the ones that I know in Japanese, so swapping one for the other is very easy. If I haven't learned to use the word in Japanese, I maintain the English keyword. It's a gradual shift that feels very natural.

Personally, I find using a Japanese keyword about equal to using an English keyword. I know, I know, AJATT AJATT English is dangerous and ruins your japanese yada yada, it isn't true however.

Using an English keyword to give a meaning of a kanji gives you what? An approximate meaning. 申 doesn't mean sayeth, but it gives you a pretty good indication of what it means. Let's say instead of sayeth, you go with 申します. What have you accomplished? It's still an approximation, you'd need 3-4 japanese words at least to get a better picture of the ACTUAL meaning of the kanji.

Trading the english keyword for a japanese one is just trading one simple association with another, it won't really give you anything. Some people will say it will teach you readings, but it only will if you add something you don't know. If you already know that 申 is used in 申します, you already know the reading and the word, you're not learning anything by using it as a keyword.

So personally, I think using Japanese keywords is worth it only if your Japanese vocabulary is very strong before you start with RTK1, and use Japanese keywords from the start. Going back and changing it won't really change anything, possibly if you are mixing up English keywords and think it will be easier with Japanese ones.