poblequadrat Wrote:The first time I quit studying Japanese was because it is basically impossible to learn Japanese vocabulary without kanji, and without SRS I ran into a wall after 50 kanji or so.
I think Heisig's premise isn't necessarily different from yours, it's just that he wants you to get all kanji down so you can naturally learn whatever vocabulary you encounter according to your needs (which, by the way, isn't quite the Core deck approach). Basically, the point of Heisig is allowing you to focus entirely on grammar and conversation. However, this means that Heisig isn't very practical if you don't do it fast, as I said in my previous post. On the other hand, knowing some basic Japanese makes Heisig much easier - you can replace some confusing keywords with actual Japanese words, and entirely skip some stories.

I'll also add that the kanji in the words Zgarbas mentions aren't particularly rare, are they? Well, I don't know what 董 is, but as for the others...

Heisig's premise isn't so much to "learn ALL the Kanji up front", it's to learn the most common ones (to familiarize you with Kanji, and give you a way to easily learn any other Kanji you'll need, later on).

He just never bothered making a perfect list of the most common ones (he does explain why: he didn't think it was important to make the perfect list, he decided it would be better to just go with the official list, plus Kanji that are needed because they show up as primitives; he strikes me as the type of person hardcore studying sessions come easily to, so he didn't see the extra effort as a big deal). He also didn't have Rikaisama, and the many other tools we have, that make reading Japanese text with the occasional unknown Kanji in it, pretty easy. I don't even think they had cut paste back then. Did they even have scissors, to physically cut and paste the piece of paper the Kanji is written on? Probably not.

So what's wrong with making his list a little better? It certainly doesn't affect the method. Do you really need to do 2000+ Kanji up front? Wouldn't learning 1500-1600 (as long as all the common ones are in there, and the primitive structure is kept intact) achieve the same purpose?

And that's not motivated by being a perfectionist, and nit picking about a few Kanji here and there. If you can shave off what looks like a pretty decent sized list in OP's link, without getting rid of anything particularly important, that's a big improvement well worth bothering with (especially if it's someone just handing you a list with a bunch of deletable Kanji on it, so you don't even have to do all that much).

P.S. In case someone hasn't heard about this, and wants to really make a dent in the size of their Kanji deck, there's also an RtK Light list floating around, that gets rid of about 1000 Kanji out of the 2000 (and, I believe, keeps all the necessary primitives).

cophnia61 Wrote:What if we take the same premise from another point of view? If RtK is essential in order to learn vocabulary, what about the other common kanji you still don't know after you do RtK vol. 1? And maybe even after RtK vol. 3?
Unfortunately even when you finish RtK 1 you will still encounter new kanji from time to time, maybe after RtK 1 you will know relatively uncommon kanji and still don't know more common ones? *

Obviously if you already know them it doesn't make sense to suspend them, but if you haven't studied them then you can by all means put RtK on pause (just review the kanji you know) and study a lot of vocabs which use the kanji you already know.


* This is based on the 4th edition, maybe in the 6th edition of RtK1 are included all the common ones you don't see in the 4th edition?

The differences between editions are minor. There are lots of common Kanji, including VERY common ones, that don't show up in any edition of RtK1.

If you do RtK3 as well, then you probably don't have to worry about learning new Kanji for a long long time. That much is true. But that's not really the reason for the RtK method. It's not like it's difficult to pick up a few new Kanji, while you're also learning vocab.

What would be difficult is trying to do vocab without knowing any, or most of the Kanji in them. That's the problem the RtK method is meant to solve. Once you know most Kanji, a few unknown ones here and there aren't a problem.

Stansfield123 Wrote:So what's wrong with making his list a little better? It certainly doesn't affect the method. Do you really need to do 2000+ Kanji up front? Wouldn't learning 1500-1600 (as long as all the common ones are in there, and the primitive structure is kept intact) achieve the same purpose?

What Stansfield wrote here made me realize a better way of what I've been trying to say...

If you were to graph a kanji frequency list, you'd see a smooth hockey stick line going from frequent, to not so frequent. But there's never any point on the line where you would want to say "these on the left are useful. these on the right are useless." It's a continuum. A gradient. So whether you start with 1000, 1600, 2000, 2200, or whatever is a somewhat arbitrary boundary. There's always going to be more kanji to learn.

I will say this.

Almost every Japanese person writes 曇る as 雲る when they write by hand. It is kind of stupid that those Kanji are even differentiated.