45000+ of which are irrelevant to everyday life.
If experienced readers really needed to look at each individual radical in each individual kanji, rather than just apply basic pattern recognition and intuitively resolve ambiguity through context, kanji would be completely dysfunctional as a script. Will they need to carefully look at a kanji they are not overly familiar with from time to time? Sure. Will they need to do that for most of the kanji in any given text? Honestly, I doubt it.
You are confusing things here. To begin with knowing radicals is how you understand also kanji that you don’t recognize. So saying they are not common, is irrelevant to the discussion. The fact is, Japanese has a ton of kanji. And just recognizing them by sight, is just fantasy. You’ll use the radicals to make assumptions about meaning.
The second part is that you will recognize a lot of kanji just on sight. But that only happens for really simple kanji, or AFTER you’ve been using the radicals as a support for your memory. And as soon as you end up being uncertain about a kanji, you’ll have to fall back on reading and understanding the radials for comprehension.
So, you’ll be needing radicals even after burning some kanji, as they simply are very visually similar to several other kanji. Radicals is just how you differentiate them.
I’m interested actually, what is the correlation between wanikani levels and radical preference?
You might not need them for the less complex kanji at the start, but when the kanji gets more and more complicated and similar, they are a blessing sent from heaven.
I do not disagree with most of this: radicals are useful to memorize kanji and to distinguish them while you are not yet fluid, and to approach kanji you do not know.
However, I uphold that a fluent reader will be able to distinguish most of the kanji in any given text - not just the simple ones - without actively needing to analyze the radicals and digging through their memory. The key here is that having “burnt” a kanji is not the same level of familiarity a fluent reader will have.
And yes, there are 45k+ kanji that most Japanese people have never even seen. They will occasionally see one of them and then they need their radicals. That is true. However, in any real text, these kanji are bound to be really, really rare.
And if a reader really needs to actively look at individual radicals within more than, say, 5% of the kanji in any given text, I cannot imagine they’d reach any kind of reasonable reading speed.
When you are talking about people who are still in the process of memorizing fairly common kanji, you are talking about people who are by definition not fluent.
So I agree that it is impossible for a learner to just distinguish kanji at a glance, and that it is impossible for anybody to know all kanji, but I disagree that a fluent reader needs to rely on knowing and analying radicals when reading an average text. That doesn’t make sense.
For reference, here is a study according to which fluent readers can achieve reading speeds of ~120 words / minute in mixed kanji / kana script: https://core.ac.uk/download/pdf/82696105.pdf
Such a speed would not be attainable for anybody who needs to pay close attention to individual radicals within any given kanji, similar to how you couldn’t reach comparable speeds in reading Latin script if you were actually reading individual letters (i. e. the experience of anybody reading a language they don’t actually speak).
I’d still say that you’re slightly missing my point. You’re hedging your comment and that’s because you cannot completely do without the radicals. Be it rare or not.
I think about it like this. There are people with absolute pitch. This is an ability you possibly can learn if born with the right genes. But that being said, unless you receive musical training and theoretical concepts for how to discern sound - you’ll never acquire absolute pitch. It’s not magic or something you are born with. It’s a learned ability in how to perceive, interpret and distinguish sound.
And that is what radicals are. They are the theoretical concepts that help us parse complex visual information in kanji. To say that you are not relying on them, even when you recognize a kanji on sight, is just not true. You are in fact using the radicals as part how to perceive, interpret and distinguish the kanji.
Whether you’re conscious about this process or not is irrelevant. It’s still part of the visual information you’re using for recognition.
So, being fluid is really only about being conscious or not about this process of recognition. That doesn’t change that you are using radicals as part of the theoretical framework for visual recognition.
I am not a neurologist but my understanding is that you can, in fact, recognize a complex shape without subconsciously chopping it up into its constituent parts, even if you did at one point rely on chopping it up before you became familiar with the more complex shape. So I feel like a more apt metaphor would be training wheels on a bycicle. They’re useful, they help you not smash your face into the gravel, but eventually you can and will get rid of them. Or maybe they’re like scaffolding, which gets removed when you’ve completed the building.
I mean, when I look at your avatar, I see BB-8 with cat ears, not a circle on top of another circle.
But I understand what you mean now.
I’d honestly say more like 49 000+. Yeah, there’s 2000 Joyo kanji, but I’d be astonished if you’d encounter more than half of those in day-to-day life.
Well, I mean, if “day-to-day” life includes things like interacting with people who have names written in kanji and reading the newspaper, it seems like you’d hit 1000 kanji in a day pretty quickly, with plenty of time left over.
80 % useful 20% not so useful. Especially if it is name totally different than the kanji.
Yeah, the 80-20 rule applies here. Although I’m pretty sure I’ve read somewhere that newspapers generally use ~3000 kanji and I think newspaper are a pretty good yardstick for everyday usage.
I guess when it comes to the fundamentals of writing the kanji characters, the radicals do help. Let’s say for an example, the fire kanji. It contains the two “drops” and a “person.” If I already practiced myself by writing the radicals, then it would be no problem for me to write the kanji and somehow makes it easier to be familiar with the look of the kanji. (At least that’s my point of view on it)
That’s a good question. Like I said above, for me around level 45 the additional radicals were just easier to remember as kanji but I’m sure that phenomenon can start earlier.
Looking over the radical list, from 45-60 there are only 6 radicals listed that aren’t Kanji themselves.
But, and this is important, all the radicals learned up until that point are going to be useful forever. Because you can look up Kanji on Jisho using them and they can help you categorize new Kanji that use them.
Get a life
You’ll need that leftover time for the handwritten menus.
Learn some manners.
I will say that I am currently using another resource in addition to WK, which is introducing Kanji to me like “Hey, this is the __ Kanji, okay now lets use it”, and it is much more likely that I mix them up, forget which is pronounced how, and confuse Kanji either because they look similar or because I learned them at a similar time. Learning simultaneously with the radicals and without has absolutely sold me on radicals.
They’re pretty dang handy, both for picking out the differences in kanji that look similar (“that one has droplets to the side and that one doesn’t, so…”) but also for searching for kanji via radicals on jisho.org or similar. Sometimes the component parts of a character blend into each other, so knowing the ways they tend to be separated up is really useful.
Radicals aren’t useless, but some of them are annoying to me mostly because they don’t appear in radical selections in online dictionaries. I can’t tell you how many times I’m on Jisho.org trying to find a radical like lip ring 可 or city 市 only to have to break them down further into their component shapes.
It’s worth noting that the way Jisho breaks down kanji into components (well, really, the databases that Jisho uses for kanji, like RADKFILE) isn’t particularly standardized either.
I still find WaniKani useful because it’s the most reliable out of all the resources that I found. It has a very straightforward and structured system. I’ve tried apps and whatnot in Chinese, but they’re not as well-organized. It’s like learning Portuguese words as a Spanish speaker. You’ll find a lot of similarities in the vocabulary (i.e. kango in Japanese), but there are still differences that should be addressed with a systematic method of learning.