I think its a tough one because you don’t know what you don’t know. That is to say, if you don’t pick up a certain feeling from a sentence and just take it at face value, often times you wouldn’t really know that there was more to it to begin with unless it had plot relevance.
From my perspective, I’ve come to pick up a lot more nuance and I feel pretty confident in my ability to notice it. Whenever a nuanced part is emphasized with the little dashed thingies, usually I have no problem understanding why that part was emphasized and what the nuance is getting at, and I can’t remember the last time a nuanced line had plot relevance and I wasn’t able to pick up on it.
So overall, I guess I’d say while definitely not native level, my ability to pick up on nuance is on par with my general reading ability and doesn’t really lag behind.
Hmm, I dunno because I haven’t actually read a book in english in…maybe 6 or so years? I also don’t really watch shows in english anymore so I would have to draw comparisons from japanese books to everyday english which is a bit biased. In general though, comedy became something I felt a lot more appreciative of as I progressed. Usually its a more manzai tsukkomi style, so not sarcasm, but I feel like both are pretty clear and weren’t too tough. Overall, I didn’t feel like the whole sarcasm thing was too much of an issue though. I had watched like 100+ anime series before reading though so maybe that helped me go in being familiarized with the kinds of comedic interactions I would be seeing in light novels.
Hmm, I wish I could give a good reply on this but honestly I haven’t really read any non standard dialects. I mean, some characters talk a bit funny based on their age/character like old man rom from re zero and stuff, and they are harder to read, but not hard, yknow. Haven’t had to read kansai ben and stuff to any large extent though.
No worries I didn’t mean to highjack your thread. I just briefly answered to a claim you did I didn’t resonate with, I believe it’s in the nature of a forum. But like I said I’m perfectly fine with keeping it this way, without going into an endless debate which is pointless. And I think I won’t write any 60 level post but we’ll see
On the other hand I absolutely agree with what you stated about reading and understanding what you are reading. And koohi it’s a great tool. I’m trying to use it more consistently and surely after level 60 I’ll try to do that in a massive way along the remaining kanji
Vanilla, you rock. Thank you so much for your reply; I instantly bookmarked it
Your answer got me thinking quite a bit. Interestingly, I am already taking a lot of time contemplating each new lesson. I also reflect on each mistake I make, and even when I am getting a card right, I often make sure that the mnemonic I use is stable enough in my mind.
Buuut my accuracy is much closer to 80% (learning + young + mature cards)
I have three hypotheses here.
Despite having been using only Anki for the last 8 months, most cards I have learned come from WK. When I study those cards, I stick with WK’s mnemonics. That violates the two ‘rules’ you mentioned (using the first thing that comes to mind, and consistent sound/idea association.)
I don’t immerse enough, esp. not into reading. That’s the trickiest part for me: I live in Japan, but haven’t much interest in the Japanese culture (outside of food).
My memory isn’t that great, at least for language?
Regarding WK cards, I have just stopped using them as my primary source for cards. Building my own mnemonics should somewhat help increase my accuracy. We will see.
Regarding immersion, I am still looking for stuff I really have fun with. But hell, I like American series more than anything.
Regarding memory, there isn’t much to do, and I am okay with that
At any rate, the time (< 1 hour) you need to learn new cards (~20) + do your review (~200) is impressively low. Despite you mentioning that focusing isn’t your general forte, your ability to keep your focus on your SRS is quite the thing! Hats off to you
I mean, I did do well in school so I would say I’m good at learning stuff, but yeah for me 20 minutes of focus doesn’t mean I do it in 20 minutes. It’s not unusual for me to do 20 reviews and be like man I should go to the store and just leave lol. Making your own mnemonics is definitely good though. If something comes to mind immediately, it has a good chance of sticking. When you usually don’t get a card, is there something in particular you usually mess up on? Meaning? Reading? Just can’t remember the mnemonic? Is it primarily older cards?
Just watched the video and while it’s a little late, way to go finishing level 60! ccookf suggested checking out your post and I appreciate the transparent honesty in Japanese’s difficulty.
If I’m understanding it correctly, how you’d suggest going about learning Japanese via books is:
Pick up something interesting to read
Use a website like Koohi.cafe to find the most common words and learn those first
Dig for sentences you know all but 1 word of and use that to learn new words.
Once you know all the words that frequently reappear, start learning the words that also appear in sequels
Once you’re proficient with that series, jump to other authors and genres to learn a wider ray of vocab
Also, a big talking point was in a trade off for the convenience of WaniKani handing you a list of words to memorize, you don’t get the value of vocab in the context of sentences. If someone was going through WaniKani, would it be just as helpful reverse searching words through light novels for example sentences, or is there often still not enough context from not reading the rest of the chapter to regain that lost value from not sentence mining?
Yes, after doing something like the core 2.1k and learning basic Grammar (maybe having an idea of roughly n4).
Just as helpful? Probably not. It would probably be tough to grasp the nuance just as well that way. With that being said, its better than nothing.
The other, and bigger issue in my opinion, that I mentioned about learning words from lists was the frequency that you see those words. It makes sense to learn frequent words first within a certain work, because they’ll naturally have the most value when it comes to boosting comprehension and you’re guaranteed to have multiple example sentences.
Nowadays, there are tools to read a book where you can use yomichan on the book. I personally do this. With that being the case, “knowing the kanji” is less valuable and you can look up any unknown words in a second. So it’s important to evaluate why you want to use wanikani in the first place and what it’s offering you. At the end of the day, kanji aren’t what’s important. Words are.
What you say about kanji lists makes a lot of logical sense. To me, the convenience factor with a specifically laid out goal start to finish makes WaniKani appealing to start. Having to spend time learning to create decks takes a lot of time and work compared to a few cards. WaniKani lays out a video game-like win condition where I can just focus on retaining information and sticking to a routine as opposed to that plus finding sentences, writing mnemonics and everything else. Also being able to type back answers means no way to lie that I got it right
I’m curious, is there a way to mine sentences knowing incredibly limited Japanese? For example, I can pick up a volume of Re:Zero’s lite novel or Yotsuba& manga, but the sentences seem to all have 2 or more words I don’t recognize.
Oh wait, noticed you said learn the core 2.1k first after submitting this. Blurred not knowing how strike through works on here
Setting up my mining setup took no more than a small part of my day, and now whenever I add words it takes only a button press to generate the card.
I’d say the core 2.1k isn’t a bad idea for you then. Either way, the choice is yours.
I would just like to make the point that while it can feel like you learn a lot on wanikani (because you do), the issue is how much of that will translate into reading ability. I would go as far as to say if person A did core 2.1k and read, and person B got level 60 on wanikani, and they started at the same time…then even if person A put in half the time that person B did, I would still have my money on person A having higher reading comprehension for what they are reading.
That’s a good question! I focused on that during the last few days. By decreasing importance:
Older cards. With that said, card maturity might mostly be a catalyst for the problems I list hereafter.
Mix of kun and onyomi, and 熟語 with kunyomi only
Transitive vs intransitive verbs: I have harder time building mnemonics for that. For instance, to tackle 痛む (transitive) and 積む (intransitive), I am using cows to remember both transitivity and intransitivity, and this is getting very much mixed in my mind. I am thinking of having one card for both verb forms. I do that for grammar points that share very similar structures, which helps.
Similar words with diverging meanings: like 鉄鉱 (iron ore) and 炭鉱 (coal mine). I mean, 鉱, what is your purpose on Earth?? A solution here would also be to review all the incriminated items simultaneously.
Fuzzy understanding of the meaning: you’ve touched that point in your video when noting that bilingual dictionaries associate many English words to the same Japanese vocab, but that in most cases, there is only one underlying concept. I feel like I am drowning under unrelated meanings in those cases, and I fail to remember anything. I guess spending more time on example sentences / monolingual dict would help.
Rendaku (although during the last few months I have systematically added rendaku mnemonics, and that really helps.)
Confusion mnemonics: I have a fuzzy visual memory. I visualize both かん and けん as guys with a round-shaped hat, so I tend to mix the two mnemonics and readings a lot…
Im trying to decide what SRS outside of wanikani and bunpro will follow me for the next years. What do you think of koohi.cafe vs jpdb.io? I have wrote a longer post in bunpro explaining what I have observed until now, I will put it also here
Edit, also after watching your video I was convinced to try reading a lot sooner that what I was planning to do, in part thats why Im trying to choose what of those 2 websites to learn vocabulary from books would be better for me. Thanks a lot for your video
Looking for a SRS website to complement, and at some point to replace, wanikani.
For what I have been trying, jpdb.io I like more how they manage the decks. You just add a volume you want to learn the vocab, and all the vocabulary and kanjis you didnt learn with them will actually be added to your lessons backlog.
Also really userfull how easy they make to see how many words do you know of a certain content
Also they teach like wanikani the kanjis in the sense of using radicals to build the kanji. And if you are going to learn a vocab, first they teach you the radical, then the kanji, and then the vocab. Also in a new deck if there is a word you have already learned, they will ignore it, so you just need to focus in keep learning new words
But their reviews and lesson sucks, since it dosn’t test you if you know it or not, and instead, just ask you if you remember it correctly.
Basically, I feel like they manage the content you want to learn really well, but the review process is a big negative point for me (creator says they will add also something similar to wanikani for reviews, but it could take a long time and we dont know how it will end up being)
For the other part, koohi.cafe I dislike the management system and the stat system compared to jpdb.io , its confusing, you need to go item by item to add them to lessons, it dosnt teach kanjis with radicals or may not teach you the kanji first than the vocab, and I dont know if they know to not show you vocab you have already learned. Its really confusing so maybe I have said something wrong.
But, koohi.cafe has the strong point that the reviews are actually really similar to wanikani, asking you to type the reading and meaning. I feel like while I like less the rest of the website compared to jpdb.io , the review system could make it worth it, but not sure.
I guess one option, but not the perfect one, is to find a way to import what i learn in koohi to jpdb.io to use their stats, and make it easy to move there if one day the add typing for reviews. Not sure if this is even possible.
Other option I guess it would to just wait for jpdb to add typing, but it could end up not being even this year, so I dont think I will do this
Other one, just use koohi.cafe, and accept their imperfections
And some other ones.
I want to do the correct decision, since there is a big chance I will be trapped with it forever. This will affect me for years, maybe decades
So, opinions on people that have used more this websites? I have mainly used each one 30m to figure what they have
One thing I’ll say is I really wouldn’t worry about transitive and intransitive. It’s basically completely irrelevant and a giant bait imo alongside with ichidan and godan.
If you are inputting, you’ll get used to what word is used in what way. You won’t even think of them in pairs, and you shouldn’t. You’ll just naturally end up using the right thing because you’ll hear it that way 100% of the time and the other way 0% of the time.
I havent used jpdb to learn so I can’t really say anything about their process. On the topic of stats though, I posted about this awhile back, but I went hard for a period of time getting stats about what I was reading and using those stats to gauge how difficult a book would be. The end result was basically the %unique words known tells you really all you’re going to get along with basic intuition. Koohi offers this stat. It’s neat looking at the other stats on jpdb.io and I do so regularly, but I don’t think they are worth pondering too much to try to figure out what’s the optimal read. The second most useful stat, imo, was sentence length and you don’t need your data to look at that.
I’m horrified that I’m just now discovering this thread, but I’m looking forward to watching the entire video (it took me long enough just to read the replies!).
This resonates so strongly I can hardly contain myself. My skills in this area are constantly improving (and increasingly dependent on Japanese only resources) but I’m wondering if you or anyone here has any suggestions regarding reading printed materials?
For a variety of reasons, much of the content that I want to read is in dead-tree books. I’m getting much more efficient when reading material on my computer (and increasingly annoyed when forced to use closed-garden readers), but reading printed books is still a real slog. As with English books, I tend to just skip stuff I can’t follow, hoping additional context will make it clear later.* Often enough I go back to research what I want with computer based tools, but even entering what I’m looking up is often far more time-consuming and painful than I’d hope.
I suspect there are no good answers other than “keep reading,” but any advice at all is appreciated.
* The ultimate example of this for me was reading William Gibson’s novel The Peripheral. I was completely lost for the first ~100 pages. Then it suddenly clicked into focus and I devoured the rest of the novel. To this day, it’s the only book I can recall finishing the last page and immediately restarting on the first. The experience of it suddenly becoming clear was the oddest sensation I’ve had in a long, long time.
I have been working my way through the 深夜食堂 manga series, and some other paper books, using an app on my iPhone called Yomiwa. Although it is not 100% foolproof, it does a remarkable job at reading almost everything I aim my phone’s camera at.
It even has an integrated srs system that makes it very easy to mark words as “ask me about this later” when you look them up.
Like I said it’s not perfect (it has a hard time reading words that are in busy backgrounds) but it is helping me read dead tree material that I wouldn’t be able to tackle without it.
I do most of my reading with printed books. I think the main thing I’d suggest is making sure your dictionary lookup process is as smooth as possible, which means something with handwriting recognition for those cases when you don’t know the reading. (I use an old-school special-purpose Casio electronic dictionary, but I imagine there must be smartphone-based approaches.) I also don’t look words or readings up unless they’re either critical for understanding or they turn up frequently enough to bug me.
Yomiwa is my JE dictionary of choice (I even purchased the outlier extension). I pretty much live in it these days.
I’m embarrassed to admit that I didn’t even know it had this ability until you mentioned it!
Thanks! I’m slowly getting better at this, both handwriting recognition (I can’t read my own half the time) and 部首 based lookup.
The book I’m struggling with currently has a combination of specialist/technical vocabulary and 関西弁 (and combinations thereof) making the process especially fun. I’m slowly building up a concordance (specifically a list within Yomiwa), but some words I just can’t find for love or money.
Right now, for example, I’ve mostly parsed this sentence fragment:
I know 底辺 is the base of a triangle, and from context I know that 立辺 means the vertical side. I’m curious about the reading, though (たちへん？ たつへん？ たちべ？ りっぺ？) but can’t find any reference to those characters together other than as names.
Thanks for publishing that. I appreciate you sharing your thoughts. I’m clearly not the only one to watch the whole two hours (!!). That’s a testament to our interest in hearing from people like yourself that have achieved a level of “comfortable literacy”. (GREAT phrase, by the way!)
When I first saw the thumbnail I thought you were wearing 着物と羽織！
You initially indicated you weren’t interested in discussion, but since you have engaged a bit in the later thread perhaps you’ll allow me to share some of my notes after watching. (Apologies for reviving an old thread.)
Is it fair to summarize your main themes as follows?
Even level 60 on WK is just the beginning of a journey (it was perhaps only 20% of your own journey so far). Further, there is no final point where you will have “made it” and mastered reading Japanese. The number of books you read isn’t the point: it’s how well you can read, how much effort it takes to read, and how well you comprehend that matters.
Building vocabulary is ultimately more difficult than learning grammar. Learn grammar basics first, then research and learn unfamiliar grammatical constructs as you come across them in your reading.
Build your vocabulary efficiently. Learn vocabulary for a particular genre, author, or even an individual book first. There’s value in learning, say, the most common ~2000 words that cross all genres first (including many words not taught on WK), but you expressly recommend against just throwing yourself at the “Core-10K” Anki deck.
Don’t despair, especially when comparing yourself to others! Everyone finds it difficult for many years. People only publish their accomplishments, the amount of effort it took to get there is usually invisible (and no different than your own struggles).
Your never-ending goal should be continuous improvements in becoming “comfortably literate,” but enjoy yourself! Read what you want to read. But if you don’t enjoy the process of learning to read, there is no point in making the journey.
I agree strongly with all of that, with three small caveats:
I doubt that everyone on this forum even aspires to a near-native level of “comfortable literacy”. I suspect many just want to get to the point where they can enjoy a few simple manga, no matter how stumbling the process. Others (like myself) are more interested in conversation (spoken, email and text), and just see reading as a means to an end: The more you can read, the clearer the language becomes and the easier it becomes to use Japanese-only resources to research points of confusion.
I’m currently struggling to get through a book on Japanese carpentry tools but, to your point, it doesn’t matter because I’m thoroughly enjoying the process and I’m passionately interested in the content. I also know the next one will be at least a little easier. Ultimately, though, I don’t care if it never truly gets easy. That I’m already able to make progress at all is a reward in itself.
Many, maybe even most, WK users do need to lower their expectations quite a bit, but there are at least a few that fully appreciate just how gargantuan the task of learning a language really is.
Forgive me, but “comfortable literacy across all genres” is admirable optimism but not practically achievable. The English equivalent would be one person “comfortable literate” reading Joyce, Chaucer, Gene Wolfe, and academic treatises in many unrelated domains. Any language has a lot more words than any one person can possibly know! Go where your interests take you, but as you point out, no-one should ever expect to stop learning new words (or forgetting old ones).
Everyone has different interests, and content comes in many forms. Manga, light novels, visual novels, games, and even full novels aren’t for everyone. I can only speak for myself, but these days I’m far more interested in browsing Japanese websites and watching Japanese tv/videos and movies than I am in reading books in any language. Learning to read even the little I already know from WK has allowed me to understand (and research) much more of the content I consume (written or spoken).
I am fortunate enough to have discovered two genres of written content that does interest me, however: senryu and Japanese carpentry books. I mention it because the latter has already made several YouTube channels much easier to understand, and even things I’ve picked up from the poetry has paid off in everyday conversation!
It’s a good video and a more honest overview of the process than one usually sees.
I do have some curiosities. You describe WK as sacrificing speed for structure.
And I agree that a lot of the more efficient methods for building vocabulary and learning kanji are also decidedly unstructured, and they’re pretty much an Ikea piece of furniture delivered in the box, with no instructions, and you’re not sure if you even have all of the parts. You just start building and making mistakes until you have something a little wobbly but just stable enough to sit down on.
Whereas if you had the instructions, they have overly complex diagrams and explanations for all of the parts. Eventually you will put it all together and it will be pretty sturdy, but you won’t get it put together any faster than if you had just started experimenting on your own. If this analogy works.
Do you think that a structurally sound Japanese language instructional that doesn’t sacrifice efficiency, doesn’t skimp on vocabulary and doesn’t require multiple other additional resources to get over the initial fluency hurdle is possible? And if so, what do you think it would look like?
No problem. I was kinda expecting for it to randomly get bumped through people seeing it through my profile and stuff.
Yeah, to be nitpicky i think the exact number was like 14% of my journey up until that point time wise. But yeah I think you got it for the most part. Don’t get tied up in the idea of finishing something meaning anything. The time you spend with a language and the quality of that time are the real predictors. The language gains don’t come from getting to the end of a book/movie/manga, it comes from what you did up until that point.
Yeah, with the note that difficulty is really just “time required”. Conceptually, grammar is much harder to grasp. But the grammar contained in say, 50 books, is absolutely dwarfed by the amount of vocab. So in terms of what will bottleneck and pose issues to you for a majority of your journey, vocab is likely to be the problem.
Couldn’t have said it better myself
Yep! I think its very productive to use others as a sort of example, or role model, in deciding how you should study and what good study habits look like, but comparing your achievements should be done with caution. If you’re gonna compare your achievements (which I do sometimes), I think that you should compare yourself with people who seem to achieve more than you and ask yourself what you can learn from them. Expecting and trying to perfectly emulate them and their success is unhealthy, but I think its productive to look at it in terms of “what can I learn from them and use in my own studies”.
I totally agree, so hopefully that was clear. If I had to say, there are probably only a few people on here who really care about getting to near native level literacy and have a similar mindset as me while doing japanese as a hobby. I tried to make it clear in my post and in the video that it was intended for people of that mindset who want to read a lot of novels and have japanese as a hobby. I tried to frame it like “people who aren’t of that mindset, keep in mind this stuff may not apply to you”.
I’m sure there are a handful of people with reasonable expectations, for sure. But I guess, as a minor rebuttal I would almost expect someone who claims to know how hard learning a language is to probably have incorrect expectations if they are a native english speaker for example. Learning japanese for us is a lot harder than learning german or french, for example. So unless they have learned a language of a similar difficulty to a very high level, I wouldn’t expect their past experiences to make them perfectly aware necessarily. At the end of the day, the only way to know exactly what it takes is to do it, right? So everyone is gonna be off the mark at least a little bit when it comes to say…what 5000 hours of study will look exactly like, and if I had to guess I’d say 99% of people would overestimate it. I don’t doubt a lot of people are a LOT closer to the mark than I was, though.
haha, well to be fair I think I explicitly said “this is for people who want to read a lot of novels” in the video iirc, so I mean yeah its not gonna cover some people like you.
I probably could have done a better job of explaining “I’m only saying this stuff about this very specific group”, since I think all your points are valid and I never meant to suggest otherwise. I just didn’t want to say “for people wanting to get to near native level and read a lot of light novels who are learning japanese as a hobby…” before every statement so I gave the blanket statement at the start. I also guess I shouldn’t have just said “high” level of reading, since “high level”'s definition varies person to person. I know some people would consider passing N1 a high level of reading, but if what I took yesterday wasn’t an out of season april fool’s joke, you could pass the first two sections with what I would consider upper beginner/lower intermediate. I’m aware my scale is very skewed compared to the norm, so using general descriptors like that was probably a poor decision.
I think your analogy isn’t a bad one, but kinda goes past an important point. If you assemble something yourself through trial and error, and especially watching pros, you’ll likely develop a deeper understanding and intuition of the way things work together. You’ll be thinking about the steps that come next, how pieces fit together, what role something might have based on its shape, etc. whereas someone following instructions has to do none of that. The person following the instructions will probably finish faster realistically. Maybe twice as fast.
The point is, in my opinion, is take someone who has built 5 tables on their own and someone who has built 10 tables while following a guide. Throw them in a room with pieces to build a new table theyve never seen and ask them to build it on their own. The ten table builder has built more tables and the amount of experience (time) they have is the same, but I would expect the 5 table builder to do a quicker and more accurate job. He has honed his intuition and understanding with his time. 10 table builder is familiar with the parts, but his assembly has taken place under the guidance of the assembly guide and his sheltered experience has prevented him from ever having to think critically about the process those parts are utilized in. Take away the crutch and he falls flat.
This is the comparison I tried to illustrate when I talked about a person learning n words and immersing for x hours probably having a higher reading level than someone who did wanikani and got to level 60 after the same amount of time.
Like something outside of immersion? Hmm yeah I think its possible, but idk if it will ever exist. I think that the problem is that it just would not be interesting lol. I think the most efficient way to do something out of context is a PERFECTLY curated set of sentences, like probably 1000s, from actual media. They are introduced in an exact order, given surrounding context when necessary, have explanations and a ton of similar sentences when there is one you don’t understand, and audio and visuals to go along with it. It would basically just be a perfectly constructed artificial immersion environment. Seems near impossible to make, though,