Greetings fellow language learners.

I'm still studying Japanese now, and will continue doing this for the rest of the year. Around the end of this year I'll start on mandarin.
However.
I was toying with the idea of switching over to Japanese keywords. So I was wondering if there are people here who did that, and then at a later time started learning Mandarin.
How did it affect your studies? What did you do with your kanji reviews?
Kept on reviewing the Japanese keywords, ditched them, changed back to English, or switch to mandarin keywords, ...?



And on a side note.
Apparently, quite a few people here start over with RTH/RSH when they start doing mandarin. Isn't it way more productive to just do RTK3, and replace the Japanese and simplified forms in RTK1 with traditional forms (or vice versa)? Or are the kanji in RTK3 quite Japanese-specific?
It's just that I wasn't planning on doing RTK3 in the context of Japanese, but perhaps if it pays off in Mandarin, I might do it now, during my Japanese studies.

Anyway, thanks for the potential future replies, random strangers.

Jorre

About side note:
I think Katsuo did a spreadsheet (yes another one) where he pairs the difference in keywords in RTH from RTK (but only for volume one I guess). You should consult that.
I don't know if Japanese keywords would hinder your Mandarin learning or not, but doing RTK3 really seems an overkill, since RTH vol. 2 is soon to be available, and I'm guessing the content is quite different from RTK3 (made from different lists for different purposes).
If you don't want to wait, I think you're better off making your own keywords from a frequency list. But that's just my opinion.

What makes you guess that the Japanese usage of characters differs so much from the Chinese? I mean, in my eyes, the point of Heisig's method is not to gain exact knowledge of an exact bounded set of Characters, or knowing what they mean... It's way more general than that.
So if you learn a few hundred characters more or less, or if the keywords differ a bit, it doesn't matter. If I never would have learned the character 襲, or if I learned it as "offense" instead of "attack", the method have worked just as well.
So it's just a matter of efficiency. Of course, you COULD learn 3000 characters all over again, knowing that at least 70% of the work is redundant and useless. Or you could just learn the 1000 extra in RTK3 and make peace with the fact that maybe you know a bit less frequently used characters, with keywords that might be a bit off, but at least you did 100% novel work, assuming that those 1000 in RTK3 are actually reasonably frequently used in Mandarin...

Somebody should show RTK3 to somebody from Taiwan/HK and ask how useful the characters in there actually are. Or compare the content with stuff like Harbaugh's book or other resources.

Anyway, thanks for the quick reply.

Jorre

Edit: A quick look at the spreadsheet from here: http://forum.koohii.com/viewtopic.php?id=7263
shows that there are 110 equivalent characters (meaning simplifications included) in RTH1 alone treated in RTK3. That's 10% down. 1260 are treated in RTK1.
So it wouldn't seem far-fetched that a lot of the characters in the upcoming RTH2/RSH2 can be found in RTK1 and RTK3.

jorrebenst Wrote:What makes you guess that the Japanese usage of characters differs so much from the Chinese?

I don't see why it should be the same either (different vocab, different syntax, different background; that a same symbol is choosen to be used in the same way or as frequently isn't obvious to me).
I'm not talking usage like in "different keyword for same character", but lists made with different characters.
You see, RTK3 is famous for including a lot of kanji about philosophy/spirituality (Heisig's original field). Plus it was made more than 20 years ago, and for another language, so even if a priori there should be a lot of common characters, maybe the similarity doesn't exceeds 50%, and for 50% or below, it's more efficient to do it yourself with a frequency list, since you already know the method.
Except if you had planned to do RTK3 for your Japanese studies anyway, in that case, I see no harm.
I mean, of course RTK3 kanji are Chinese symbols, but are they useful symbols, I just don't know.

EratiK Wrote:You see, RTK3 is famous for including a lot of kanji about philosophy/spirituality (Heisig's original field).

I hear this a lot, but could you elaborate this please? when I skim through the reading section of RTK3 and look at the compounds used, I see the occasional Buddhist term, but most of what I superficially skim are actually some pretty normal words, like "lunar eclipse" or "whip", stuff like that.
Are there some numbers on this somewhere?

EratiK Wrote:Plus it was made more than 20 years ago, and for another language, so even if a priori there should be a lot of common characters, maybe the similarity doesn't exceeds 50%, and for 50% or below, ...

Well I don't know about that, since RTK 1 is even older than RTK3, also for a different language, and 85% of the characters in RTH are found in RTK1... And another 7 percent is in RTK3, so 92% of the characters in RTH1 are found in RTK1+3. Why would the number for RTH2 suddenly drop to near-useless levels?


I'm not even sure I want to make more stories anyway, right now I'm more thinking in the lines of just learning everything in context.
I'm definitely not going through the trouble of making my own list of characters to learn. It's either precompiled, or none at all.

Haha, so much for a "side note", lol.

Cheers

Jorre

jorrebenst Wrote:I'm definitely not going through the trouble of making my own list of characters to learn. It's either precompiled, or none at all.
Jorre

You don't have to necessarily. There are many frequency based character lists available: 3000 or 5000 characters etc. Word lists too... The problem is that they are not using single keywords.

Anyway, I randomly checked RTK3 lists with a frequency list, and half the time the character were over 5000 most useful. I wouldn't do RTK3 if I wanted to focus on Chinese.

So I've skimmed through RTK3 keywords' list, and apart from the final 5% old forms -- the buddhist related terms (like kalpa, bodhisattva) seem to be under 10%, and really weird keywords like supinate are also very few, so that makes at least 75% of the keywords user friendly (ie probably reflecting useful kanji).

But still, I'm not sure that prooves anything regarding Mandarin. The numbers in similarity are likely to drop, because after 2000, Heisig explains somewhere the frequency is based on something like 10 times less frequent items, so it all becomes a matter of corpus, and since the corpus is different for RTH...

I remember someone making a list of all the keywords that where seen in RTK vs seen in RTH. Maybe someone could make a deck of the differences and people would only have to do the difference, not the whole thing again but for chinese.

Only seems to strengthen my viewpoint that any additional formal character study is not beneficial for people coming from RTK1, and that learning from context is probably the best way (ie pretty much the same as learning from frequency lists).

So any people with japanese keyword experience to chime in?

Jorre

To make going through all of RTH even though I've done RTK a little less useless in my eyes (though I think it's useful to begin with), I've included the reading on my cards (with audio) as well as make it a point to learn the reading as I learn the character. I usually work the reading into my story, or if it's obvious (due to radicals) I don't bother.

Why not try that?

Ryuujin27 Wrote:To make going through all of RTH even though I've done RTK a little less useless in my eyes (though I think it's useful to begin with), I've included the reading on my cards (with audio) as well as make it a point to learn the reading as I learn the character. I usually work the reading into my story, or if it's obvious (due to radicals) I don't bother.

Why not try that?

Might as well learn them in a sentence then. How much time do you spend on going through 3000 characters, making up a story for each one of them, reviewing them all? Unless I'm really mistaken about the idea that RTK1 is enough "luggage" to start on sentences, I'd rather do that.
I spent 9 months on RTK1. It wasn't 9 months of kanji only, as I got a good enough grasp of basic Japanese to start on KO2k1 immediately afterward, but still, I don't want to, can't and won't spend another 9 months of learning some weird pseudo-language consisting of 3000 random useless stories, only to learn a system that I'm already familiar with.
5000 stories man, 5000. It'll make my brains fry.

Cheers

Jorre

Edit: I don't regret doing RTK1 nor spending so much time on it, BUT, I was a total kanji noob when I started. No knowledge whatsoever. It was the greatest thing I ever did back then, and I still faithfully do my daily reviews.

6 months? It only took me 3months. But that was b/c I was so hardcore back then(i.e. kept saying, I want to know so much kanji. 3000+). Now I'm more experienced on how to learn.

Usual if people are doing RTK, one could still do sentences/basic vocab. But I agree that in the beginning one thing at a time is necessary. Since, one is not used to kanji/the writing system of the language.

jorrebenst Wrote:I spent 9 months on RTK1. It wasn't 9 months of kanji only, as I got a good enough grasp of basic Japanese to start on KO2k1 immediately afterward, but still, I don't want to, can't and won't spend another 9 months of learning some weird pseudo-language consisting of 3000 random useless stories, only to learn a system that I'm already familiar with.

Well anyway you won't since you already did RTK1.
For example, for RTH1, you already know 1260, so you'd only have 240 new kanji to learn. For the rest you won't be remaking a story (since you're already familiar with the kanji), but you'll only add another keyword (if the meaning isn't close) plus eventually the reading to make sure your mind dissociates the Japanese and the Chinese system.
So in worst case scenario, with RTH2 you'll only have like 1000 new stories to make (the others possibly being RTK1 or supplement redundant). A hundred a day = ten days ; beats 9 months.

I don't know...

I've been doing RTH on and off for the past 2 weeks and I'm almost on 1000. When I sit down I can really bang them out no problem. I've noticed a decent number of changes and a decent number of different characters from RTK, so I'm not sorry I'm doing it.

And to that note, I'm not trying to memorize the readings. I'm just trying to familiarize myself with them so they stick easier later on when I start drilling them.

Anyway, I'm still not convinced that RTH teaches anything you can't pick up from context.
So really nobody with any experience with both Japanese keywords and Mandarin?

Cheers

Jorre

jorrebenst Wrote:Anyway, I'm still not convinced that RTH teaches anything you can't pick up from context.

Can you explain what you mean by "pick up from context"? I'm not 100% sure it's that simple.

I've moved from RtK to RtH Traditional, and so far I'm finding it's like this:

- Around 50% of the hanji are the same as their japanese counterparts - same keyword, same primitives (sometimes slight variation on writing or font, but thats easy to memorise).
- Around 20% are kanji that I know from RtK, but they have a different keyword / meaning. Yes you could "pick it up from context", i.e. try and remember this japanese kanji means something different in chinese, but it is tough breaking that bond that I've built up in 2 years of daily kanji reviews.
- 15% are brand new kanji - these are super-easy to learn (just make a new story).
- Final 15% are kanji I know, but are written differently with new primitives or stroke order. These are usually quite big changes, and occur due to japanese kanji having many primitives removed when they were simplified. Some of the traditional variants are crazy hard! I basically have to make a whole new story for these ones, but it's confusing cos they have the same keyword - these are the hardest to learn.

For me I have totally switched from Japanese to Chinese, so I simply stopped reviewing RtK and started over with RtH. If I was doing both the hardest part would be different kanji/hanzi with the same keyword - is it possible to have 2 stories for the same keyword with confusion? who knows...has anyone done this?

Well, "picking up from context" sounds quite ridiculous. What could this word possibly mean, without knowing it's true meaning, just by "guessing" (through Heisig's keywords) and picking up from context: 伝統、踏切? Certainly there are more difficult compounds. But that's what Japanese and Chinese are about, kanji or hanzi in compounds and not just on their own like in Heisig's books. Heisig is basically a preparation ("Start learning Japanese after Heisig."), I think.

Tori-kun Wrote:What could this word possibly mean, without knowing it's true meaning, just by "guessing" (through Heisig's keywords) and picking up from context: 伝統、踏切?

Lol, do people actually do that even in japanese? That's exactly my point. When I'm reading a text in Japanese, I don't even think one second about what keywords the kanji have that I'm reading. If I see a word I don't know, I look it up in a dictionary, I don't "guess" the meaning based on keywords. Then I memorize this word.
So by "picking up from context" I just mean: learn words as you encounter them. who gives a crap what keywords their characters have. If you don't know one of the characters in this new word, you add a close delete card, or a heisig-style card, or whatever you think of to do writing practice.

So should I encounter "踏切" for example in some Chinese text (supposing that it's a Chinese word =P), and I have no idea what it means, and above all, suppose I have never seen the first character before, then I just look it up, and add it to my SRS or whatever studying regime I have. Spend no more thought on it. If I want to write this character, I add a writing card.

That's what I mean with "picking up from context".

Great thread necro by the way =P

Cheers

Jorre

jorrebenst Wrote:So should I encounter "踏切" for example in some Chinese text (supposing that it's a Chinese word =P), and I have no idea what it means, and above all, suppose I have never seen the first character before, then I just look it up, and add it to my SRS or whatever studying regime I have. Spend no more thought on it. If I want to write this character, I add a writing card.

When i started studying chinese, I assumed RtK would get me by fine. Maybe if I was learning simplified Chinese it would have. Problem is, I'm doing traditional, and pretty soon i went back to a pre-rtk state of "I'm just looking at a big bunch of squiggles - i cant tell them apart, let alone memorise or write them".

Remember for chinese you need at *least* 6,000 characters to read a newspaper, 3x more than Japanese. So rather than just adding a few characters here and there, it's more likely that a majority of hanzi you encounter will be unknown to you. Sure you can add them as you enounter them, but surely it's more logical to work through them systematically using a heisig like method?

I figured I may as well start RtH, to learn the characters I didn't know, and learn traditional variants of ones i did. But then hanzi.koohii launched, and since I'm giving up japanese, i figured I may as well start over and add everything again. This approach is not optimal, maybe RtK3 makes more sense if you're planning to do both Japanese and Chinese.

I've never seen any data that show that you need 6,000 characters to read a Chinese newspaper. I think it's more like 3,000, only a bit more than Japanese, from the numbers I've seen.

I started doing all of RTH, having done RTK, but decided that it wasn't worth it. I do need to find a deck with the few hundred characters that aren't in RTK1, but I haven't had the chance.

jorrebenst Wrote:Lol, do people actually do that even in japanese? That's exactly my point. When I'm reading a text in Japanese, I don't even think one second about what keywords the kanji have that I'm reading.

Of course people do! Isn't knowing a rough meaning one of the main points of doing a Heisg book? For me knowing the keywords for the hanzi is really important, as it 1) helps memorise words and compounds, and 2) is extremely useful if you live in Asia! Being able to guess a rough meaning for the hanzi allows me to read signs, posters, newspaper headings - basically survive as an adult until I actually learn the Chinese readings for them.

Of course thinking of the keywords is probably not so useful for someone at an intimidate / advanced stage, or someone who has the luxury of studying as their own pace and doesn't have to use them daily.

But anyway this topic has been discussed to death on here before!

bertoni Wrote:I've never seen any data that show that you need 6,000 characters to read a Chinese newspaper. I think it's more like 3,000, only a bit more than Japanese, from the numbers I've seen.

That's great if true - seems totally obtainable! I wonder if that includes compounds though; I mean you learn 3,000 character readings and then do you have to learn what the combinations mean separately?

I have no way to really accurately count how many characters I know. I suspect somewhere around 15,000, and if I sit down to read a newspaper, I will still miss a few characters.

And yes, we're talking individual characters.. when you begin to combine them, this number increases exponentially.

jorrebenst Wrote:Anyway, I'm still not convinced that RTH teaches anything you can't pick up from context.
So really nobody with any experience with both Japanese keywords and Mandarin?

I've started learning Mandarin after reaching a level of Japanese where study is no longer required. What I'm finding is pretty much in accordance with your statement. I'm finding myself able to learn additional kanji (traditional) and differences easily just by learning mandarin vocabulary (i don't give a crap about keywords). It really does makes me wonder if RTK was even necessary at all, but such a statement would have been unthinkable to me 3 years ago.

@zerOrange
How do know 15,000 characters? Wouldn't you have to start delving pretty deeply into archaic or classical texts to get any use out of those?

In addition to learning the characters, you need to learn the combinations.

I agree that RTH isn't useful after RTK, IME. Just reading is enough to pick up the new usages and pronuncations.

nadiatims Wrote:@zerOrange
How do know 15,000 characters? Wouldn't you have to start delving pretty deeply into archaic or classical texts to get any use out of those?

Classical, yes, archaic... that depends upon your definition. I mean, I wouldn't consider 易经 or 孙子兵法, 西游记, 三国 etc.. 'archaic'. Understanding them is central to understanding ancient Chinese culture (and to a large extent, modern culture). Certainly there are characters that I have seen once and haven't encountered since, but frequency is completely irrelevant to fluency.

I also have no reliable way to tell. I am certain that it is over, say, 11000, but could be as high as 15-16.

Anyway, putting an actual number on it is a misguided way of going about approximating how well you're doing. Instead, pick up a book/newspaper/article that you think you'd like to read. Within a few minutes it will become obvious if the amount of characters you know is sufficient - this is also very personal (some people are more or less tolerant than other people of ignorance).