Learning Kanji-readings before respective vocabulary being kinda redundant? (Sry for the wall of text)

Hey I’d have a question please which I’ve found one topic on already but I try to extend that thought with a kind of a text wall (sry for that).

I just started Wanikani a few days ago and study Japanese in general for a few weeks now, so I’m a beginner merely knowing kana and some very basic Kanji.

While I like the mnemonic concept of WK since their stories are really cool and actually helpful imo, I found myself seeing few to zero sense in getting forced to learn a (or better “any”) respective reading(s) on how the Kanji could be pronounced (or as WK stated is rather “the most common” way); what is the actual sense in that? I mean I’m getting taught 1 or even more readings of a Kanji, for example 大 is being taught as たい and だい - yeah great, but that doesn’t help me in the slightest when a Japanese says “ŌkĪ Inu (大きい犬)/ big dog” since たい and だい are both On’yomi-readings, while おお being Kun’yomi. Another example with “big”, this time without hiragana so one might think now it’s pure On’yomi-reading time, is “大人 / adult”, which is pronounced おとな more commonly instead of だいにん, so when I’d talk to a Japanese and saying “だいにん” they might propably understand what I’m referring to but when they talk “おとな” to me I’m left in the dark despite I know “the most common” way of reading 大 and 人.

If I had instead simply learned 大人 as おとな, then I automatically would have been able to see one way of the usage of how 大 and 人 can be pronounced, without directly learning the pronounciation of the “puzzle pieces” beforehand, but with having a much more pratical usage (yes WK teaches me おとな additionally which is good, but that makes the separate learning of the sole reading of 大 and 人 even more redundant in my honest opinion, even in case there will eventually come words where 大 is pronounced as だい, since then I would again automatically see that 大 can also be read as だい at the time I encounter that new vocabulary) - so in the end I would rather prefer to learn more targeted instead of “this CAN also mean this and also that and so on” as it seems this is a lot of “unnecessary” input which I could use to learn other vocabulary instead.

If I would like to order a glass of water I’d always say "水、おねがいします where 水 is pronounced Mizu instead of Sui, where Sui is primarily taught on WK - in fact I’ve seen Japanese learning apps where a Kanji indeed is shown instead of pure kana only, but it’s also neither pronounced nor expected to be memorized and the reading is merely being hinted as a “side note” in case the learner might be interested in, which I think is really helpful to not blow up the brain too much; like saying “neat to know but no compulsion”. Altogether I’m convinced the slang putting the cart before the horse would fit here for a maybe more natural way in learning the language (meaning to learn the specific vocabular first in order to see or “get a feeling” for how resp. containing Kanji can be read as, instead of the contrary way getting taught some possible readings and then going on a journey with that “insufficient” knowledge). Can someone enlighten me on how they feel about that? All this ain’t no rant btw but I really feel kinda unproductive already, despite my else big motivation in learning Japanese in general. Thx a ton.

There are 6 vocabs which use mizu for 水
There are 31 vocabs which use sui for 水

Therefore, sui is the better reading to read first. This is the same logic for pretty much all items.

On the topic of why not learn all words first:

  1. Not all readings can be easily discerned from just looking at the hiragana, then the kanji. Not only you have your usual “this is a reading for the word, not the kanji”, some irregular readings are corruptions of onyomi and kunyomi, which would complicate things.
  2. This is a kanji website for learning kanji. The vocab is here to reinforce already known readings and then teach you less common new readings

Unfortunately, you would have been incorrect if you’d made that assumption. おとな is a jukujikun reading, which is a reading that belongs only to the word as a whole, and not the individual kanji. Other similar words include 今日きょう and 田舎いなか.


I think what helps a lot when considering something like this is to remember that WaniKani is a tool for Learning Kanji, not a tool for learning the Japanese Language per se.

With that in mind, it makes sense that they are teaching you all the ways that a kanji can be pronounced, and use vocabulary to reinforce those pronunciations.


I agree that it’s redundant to a certain extent, but the trade off is that it reduces the mental effort to learn new words. By getting some exposure to the kanji (both a reading and a meaning) ahead of time, there’s less to learn when you learn words using that kanji. Yes, you have to know which reading to use for a given word (and there are complete exceptions like 大人), but guessing the correct reading becomes easier over time as you learn more. Of course, this approach doesn’t work for everyone and not everyone learns this way, but it definitely works for some people. At the very least, the first three levels of WaniKani are free, so you might as well complete all three levels and see if your view changes. :slight_smile:


since the onyomi usage is much more common with kanji compounds, it definitely makes unknown vocab easier to look up if you can guess how they’re supposed to be read.


I think it’s good to remember learning kanji isn’t really necessary to learn a fair amount of Japanese. It mostly becomes useful to know the onyomi when you’re reading, and reading is probably the best way to progress from beginner-intermediate Japanese to advanced Japanese.

Right now it’ll feel pretty useless but you’ll be ahead of the curve once you learn the basics.

You’re complaining about what sets Wanikani apart from other methods.

There are alternative approaches, such as RTK (Remembering the Kanji), where you don’t learn Kanji readings at all. Instead you’re expected to later learn them in the context of vocabulary (which is not really part of RTK itself, so you need to learn vocab separately). As far as I know, RTK also uses mnemonics.

Maybe you can try out both for a bit to see what you prefer. I only ever tried Wanikani, so I can’t really speak for it, but I think many people have learned from RTK successfully. Sometimes different people just find different approaches useful.

The trick to finish WaniKani is still, in the end, knowing the vocabularies - at least the ones that WaniKani provided, instead of just looking at Kanji and Okurigana.

I think it is expected to be a little redundant, but it isn’t a bad thing.

But then, there are a lot of Kanji compounds that WaniKani didn’t teach. WaniKani would make you better at guessing the readings.

Pretty much what @seanblue said - if you learn them WaniKani’s way, a couple of the most common readings for the kanji up front, then reinforce those with some vocabulary, then start adding some of the less-frequent readings with vocabulary, then at some point you’re going to get a feel for it and find yourself guessing a lot of readings right the first time. (You may not be sure you’re right until you look them up, but guessing a reading makes them easier to look up). You won’t even know how you do this, it just starts to “feel” right which reading is correct. As opposed to having to memorize thousands of arbitrary sounds for vocabulary words.

Even the unusual ones become that much more memorable, like 大人, because they’re oddballs.

(And some will frustrate you seemingly forever, like nin/jin or tai/dai. Sorry. Even those you can start to predict better than half the time, but you’re still going to guess wrong a lot on those.)

(Also be aware that the first few levels have a higher fraction of oddballs, which makes it seem worse than it really is.)