Wanikani's vs. a (more?) Systematic Approach

WaniKani
1

Hi there! I have been using WaniKani for about a year now and I am getting fairly serious about my Japanese study. I want to ask a question—not from a place of criticism, but to hopefully spark a genuine discussion—about pedagogy and how WaniKani approaches teaching kanji.

WaniKani seems to take an approach of talking about groupings of visual elements as “radicals”. These differ frequently from the accepted standard radicals in name and structure. Here’s an example of what I mean.

人 and ｲ are presented as different radicals in WaniKani. Because of that, you’d never consider grouping, say, 人 and 何 and 休. I assume they are split this way because they look a lot different, but from a historical context they all contain the man/person radical:

人 - a man
何 - a man carrying something
休 - a man resting against a tree

This is almost a toy example, but more abstract characters like 北 originated from an ideogrammic compound of two men standing back to back as well…but WaniKani presents it as a compound of “ice”, “stick”, and “spoon”.

I could be totally wrong, but it seems like maybe leveraging more of the historical context, using the standardized radicals, and maybe even building more of the mnemonics out of the historical origins of these symbols would better set up Japanese language learners for long-term success both in terms of interacting with native speakers and even remembering the characters and their meanings—and even readings, since kanji with common radicals often share readings.

I’m curious whether the WaniKani devs have considered this approach and what reasons if any they had for taking a different path? It seems like WaniKani is potentially setting up learners for confusion once they reach a certain level and realize that a lot of the things they thought they learned are called something else or understood in a different way by many (most?) of the native Chinese- or Japanese-speaking community. Edit: Also! I think in many ways it would greatly speed up learning and retention to group characters by their radicals, especially when they share common readings.

Thank you for reading this far if you did! I hope this doesn’t come off as unfairly critical, I’m genuinely passionate about Japanese language learning and this question is coming from a place of wanting myself and others to have the best shot at the most effective path to recognition and comprehension.

12 Likes
3

After a while radicals become a lot less important in recognising kanji because your brain learns to remember and recognise kanji as a whole. Eventually recognising the individual kanji becomes somewhat less important too, and you just sight read words.

If I ever am in a conversation with a Japanese person about radicals, and they tell me I’ve got them all wrong, I’ll be delighted, because it’ll mean my Japanese conversation skills have improved enormously.

8 Likes
4

I love this so much. :slight_smile:

I think it comes down to two schools of thought on learning kanji in particular and Japanese in general, each with their own pros and cons:

a) Academic: Some of us are interested to understand the language deeply, and have an interest in knowing the “whys”, “how’s”, etc. of how it all fits together.

Pro: For some people this makes things more logical, interesting, and memorable.
Con: This is more time consuming, and isn’t really necessary to be able to listen, speak, read, and write effectively.

b) Practical: Some folks just need/want to be able to learn to be able to use the language, and either aren’t interested in the history or reasoning, don’t have the time or energy to deal with that, or both.

Pro: Faster to gain ability and actually use the language, which for many is the only thing that matters.
Con: Potentially less sticky due to lacking context, though that may just depend on the person.

There are parallels to other languages as well, of course. I’m fascinated to learn the origins of words and phrases in English, but I definitely didn’t need any of that to be fluent, just as Japanese people who don’t go on to study language at an institution of higher education aren’t any less capable in their native language.

in the end, I’d argue that WaniKani doesn’t need to go further (in the direction of “a”) to be effective since for most people (skewing more toward “b”) it won’t matter, and those of us with the inclination can further/broaden our understanding using other resources anyway. After all, we’re learning kanji either way, and that unlocks a whole other world of materials to draw from in Japanese, in additional to what’s already available in English.

I’d love to know what resources you’ve found useful though! :pray:

P.S: As an aside, I don’t recall where or when I learned that 人 and ｲ are the same, and while I always find that interesting I can’t say that either that knowledge or the lack of it has had any positive or negative impact on my learning path. :thinking:

6 Likes
5

This will give you the why in a nutshell.

2 Likes
6

My personal take on this is that the historical/etymology side of things is often in conflict with other desirable qualities like consistency and memorability (so a more historical/tradition based approach is less systematic, not more). Sometimes in designing mnemonics you can pick things up from the traditional radical names, but sometimes it works out better not to. This I think is why most modern Western style mnemonic based systems do a lot of creating their own names for things. For example, some desirable properties of a mnemonic system include:

  • components written differently get different keywords (especially important if you want to be able to write the characters from memory)
  • keywords good for hanging mnemonic stories on, so concrete nouns like ‘triceratops’ rather than vague ones like ‘small’ or ‘again’
  • every component gets a keyword (the traditional radical system was developed for organising characters in dictionaries, so it assigns only one radical per character and doesn’t need to name every subpart of a character)
  • mnemonics that are straightforward easy to remember stories rather than confusing “this means X because the right hand part used to be written like this but it got simplified to this other thing later” etymologies

In any given western style system there’s always going to be room for improvement (over 2000 characters some inconsistencies or poor choices are bound to get in), so I don’t want to specifically defend either WK or RTK here. But I think the general approach is sound.

If you want the “why” of why kanji are the way they are, you could try Seeley, Henshall and Fan’s The Complete Guide to Japanese Kanji. This is the updated version of Henshall’s old “A Guide to Remembering Japanese Characters” book. The Amazon page has some images so you can see what you get for typical kanji, but basically they give a brief etymology of each kanji and a mnemonic phrase that typically incorporates the components and one or more of the meanings. The etymologies are trustable, in that they give references and mention when there are multiple opinions.

If you want the traditional radical names you can always learn them later - but I doubt most people will ever need to care about anything beyond the really common ones.

4 Likes
7

Pretty much sums up it all. Etymology can be something extra to help you to remember certain characters where it really makes a lot of sense and sticks, but overall definitely isn’t going to be the fastest way. I’m into memory sports by extension of doing multi blind rubiks cube solves and the systems for fastest retention basically always tend to be consistently encoding the piece of information into visually distinct nouns and have them together in some sort of story that is very memorable and really use your imagination to bring the scene to life in your mind (and potentially placing that scene at a certain location). There’s also PAO systems that use a verb and I’ve seen some people who do it for mandarin but none for japanese actually now that I think about it.

Anyways, yeah it might seem like the current mnemonic system is more roundabout and thus more inefficient, but memory is one thing where encoding things into round about and weird blocks of information actually is the more effective method. WK maybe could improve on some mnemonics and how they handle radicals that contain other radicals, but in general the system itself is pretty sound as PM said. It’s how ordinary people can do stuff like this

One thing I will add is

I learned radicals by the wk method and went on to have no problems. The average learner will really have very little interaction with them (except maybe if they get into writing or kanken?). Even talking with japanese people there aren’t any problems really. Just a sample size of 1, but absolutely zero confusion and I can explain kanji perfectly fine to japanese people and haven’t had problems understanding explanations either that I can remember.

4 Likes
8

On this subpoint, my take is “yes, but there are tradeoffs”. A good order of teaching the characters ought to pay attention to at least:

  • simpler kanji before more complex kanji
  • as you introduce each component, add the kanji that use it before moving on to the next component
  • more common kanji before rarer ones
  • teach vocab that teach and reinforce each reading of a kanji at the same time as that kanji
  • teach commonly used words before rarer ones

but unfortunately these often pull in different directions – for instance if you teach all the characters that have a given component at the same time you’re likely to teach rarer kanji early at the expense of more common and useful ones. (Given how few people make it all the way to WK60 and how long it takes, this matters quite a lot: it means you can start actually using what you’ve learned in textbooks and reading more effectively even by L15 or L30, which helps with keeping them in memory, is motivating, and is practically useful.)

RTK goes for a much more pure “kanji with same component all together” ordering, and as a result tends to get criticised for things like teaching “gall bladder” very early and “bird” very late. (Plus it doesn’t try to address the vocabulary part at all.)

2 Likes