The word “cob” is used as an English mnemonic for at least 3 different readings.
拳 こぶ＋し
零す こぼ＋す
壁 かべ
It’s counter-productive to tell learners “the mnemonic is ‘cob’” if they’ve memorized “こぶ” as the reading for “cob” and now you’re saying it’s “こぼ” or “かべ”. A 1 to 1 mapping would serve users much more effectively, though I’m sure it would make creating the mnemonics more challenging. But really, that’s what WK is for. If everyone was committed to just coming up with their own mnemonics for everything we could all just use Anki.
I think this is something they tried to improve in the recent overhaul and want to improve in general, so you might want to consider emailing them about this.
Personally I use ‘cubby’ for かべ, though it’s a bit of a stretch
If it’s something you want changed, you can email the team at hello@wanikani.com. Bare in mind, they like it when people have a suggestion for a replacement for the radical/mnemonic. Maybe provide a few examples of your substitute in the words that bother you most to show them how your substitute could be better than what’s there currently. If you do that, they’ll consider your suggestion more seriously, as long as it benefits the whole population of WK. Best of luck
I agree that this is a problem. That’s one of the reasons why I do make my own mnemonics for all the kanji. I think the reasoning for the inconsistency is that the mnemonic doesn’t have to be a perfect match as long as it jogs your memory. But it really doesn’t help, if you can’t tell which memory it’s supposed to jog. Was it かべ, こぶ, or こぼ?
It would be best if WaniKani used one English sound for each Japanese mora. Though that would in some cases make writing coherent mnemonics more difficult. Also, it would be nice if the people these mnemonics were based on weren’t video game characters that I’ve never heard of, but that’s a complaint for another day.
It’s also worth considering that at a certain point it’s deemed you’ll know a kanji well enough to not have to remember the mnemonic, you’ll just be able to recall the correct meaning/reading by looking at the kanji, and it’ll be near-instant recognition. So I’m not sure how much of a problem it really is other than a short term frustration that may mean you fail the kanji a couple more times. More failure = more attention. More attention = SRS drills it into you.
I think it is a problem because its the kind of thing that can infect your knowledge when you first learn it. It’s not a new thing that just comes in after you know a kanji. Its like if you first learn to pronounce something wrong, its harder to fix it later on when it’s been reinforced through time. If you start right off the bat being confused if a reading should be KOTA or KOTO because Wanikani has used a coat mnemonic for both of those readings, then as hard has you try, there will always be that doubt in the moment about which one it is, so you actually never do internalize the reading without using the mnemonic, because the doubt prevents you from ever moving past the mnemonic.
Just speaking from experience here. Not sure if others can relate. I think a good number of leaches hatched from those eggs before I caught on and started making up my own visualizations.
No, I totally can relate. I think they used to use ‘honey’ for both ほね and はね, and now I can never remember which is which between wing and bone. You learn one, then later learn the other, and then when the first one pops up again you get it wrong because you know the mnemonic and from then on it’s just confusion all the way (often it takes a while just to realise the problem to be able to fix it).
But they’ve evidently changed the reading mnemonic for ほね since, and this is definitely the kind of thing they want to improve.
