I think ‘rewarding memorization’ doesn’t mean the same thing as ‘impossible for someone who hasn’t crammed’ So we probably agree. I do think the structure of the jlpt specifically, as something which does not measure output by design, does reward memorization of facts. I don’t think that’s a problem though, since theoretically we all know that going in. I would think the bigger mistake is to think that the jlpt measures fluency. Similar to thinking IQ tests measure your actual, practical intelligence. I also think that’s a problem some beginners have so hopefully this conversation helps them both ways.
Edit: the N1 though must be easier than I thought if it’s trivial. I was comparing it to the English test. I watched a video of someone taking it (not native) with some of my friends, and about half of us (I double majored in English) thought it was doable and the other half thought it was absurd. We came to the conclusion that it was “fun but unnecessary” and I assumed N1 would be similar. To hear that it is “trivial” means my assumption that the tests are similar was wrong. That’s fascinating!
I was specifically referring to the quote about if cramming Japanese content would give you a good score on the jlpt. I think you’re talking about learning Japanese as a whole. I don’t think those are the same thing, really.
Sorry I’m bad at this on mobile, but the exact quote I was referring to is this:
I think the way memorization results in strong short term memory and ability to recall rote data would result in a better JLPT score. That was my op. I think most structured exams reward memorization, actually. It’s not a dig at the JLPT; it’s a comment on how people tend to study to maximize exam scores.
Ahyway I hope that clears my points up for everyone, I was specifically talking about outcomes from cramming for the JLPT, not learning Japanese as a concept.
Fair enough. There’s no pressure, I just really wanted to see someone else’s reaction and it had been mentioned in that thread. If you don’t find it interesting, then it isn’t worth reading.
When you studied the language you were studying before this (I forget if you said it was German?) how far did you get? Did you take an exam? If so what did you score and do you think you would be able to get the same score without studying for it again now?
I think your brain (or anyone who can really cram material tbh) works very differently from mine I find it interesting.
Sorry for editing again:
I just realised that, Homa, if you thought I was talking to you and not making a hypothesis on the question about JLPT cramming, you probably assumed I was talking about your speedrunning. I was not trying to claim anything about your study habits just like I didn’t in the other thread.
I truly, deeply believe that everyone learns differently, and since my brain doesn’t work the way a normal person’s does, I have no business telling anyone else how they should study. I know you have a whole plan, since I saw you make it in that thread, and that is totally fine by me. I can’t memorize things well (I’m the worst at it I’ve ever met, I think ) but that doesn’t mean I think other people should do things differently; they aren’t me.
Trivial if you have a high level of Japanese. It is absolutely trivial for a native adult though. I teach in a high school in Japan and sometimes share N1 questions with students. 99% of the time they can instantly answer. The last 1% where they pause or make an error is due to bad test taking practices normally (not reading the question properly, etc). On occasion it is possible to find a question that people get wrong but it’s as unlikely as a native English speaker using the wrong preposition.
Which English test are you talking about? JLPT N1 is relatively similar in style and level (not exactly but close enough) to 英検 一級 although it has some output as well. You can find practice papers online easily if you want to have a look.
On the CEFR scale I would say an N1 pass (100/180) is around a solid B2 level of comprehension. You could be C1 or C2 and take the test, of course. This also obviously excludes output. Next year they’re planning on introducing a CEFR equivalency scale to JLPT results so this should hopefully put to rest the kind of discussion we’re having now.
JLPT N1 is trivial in the sense that if you passed with a score of like 110 you have essentially trivialised the test because there is no need to take the test anymore because you “passed” and are unlikely to fail it again. I wonder what the pass rate would be if say the necessary passing score was 50/50/50 for each section with an overall requirement of 160? Would people say it was trivial at that point? I might even get a bit nervous taking the N3 if the passing score was that high. But it doesn’t matter anyway because no one cares about the arbitrary difference between getting a 100 and 180 in real life. The only place I have seen specific JLPT requirements are on ENGLISH language job listings ironically enough. Oh and online English language Japanese learning forums - hehe
Not a dig at anyone here but argument about how difficult and how long the it takes to pass JLPT have been going on for years and years, and is usually coming from people who have either passed the test and thus have absolutely no reason to take the test again or people who have never taken the test and don’t understand the flaws of the JLPT itself are not in passing it or your score, but more deeply rooted in its content (dry, useless texts about insects and dinosaurs, and an unrealistic scripted listening section) as well as execution (substandard test conditions, using CDs and speakers, and pencil only multiple choice test in the year 2024).
Thank you for this pointer (and in general for insightful posts as usual).
Some of the vocabulary in section 1 was unfamiliar to my friends who are not native, but live in English-speaking environment and work in software development. To the point that a couple of questions would have to be guessed. But I estimate they’d still comfortably get an overall pass.
The test doesn’t seem to have a dedicated grammar section. I wonder if it’s due to difference in philosophy behind the test (e.g. if you understand the texts, you should have understood enough grammar), or difference between languages (more unintuitive grammar in Japanese).
I studied English in school + tree long and learning it with textbooks (which I can not take anymore), and then become quite fluent watching YouTube and writing stuff here and there.
The second language was Italian, and I am still very bad, did not do any tests. I can read Harry Potter (and methods of rationality) with lingq and that is basically it, I had no idea how to learn languages when started that one, it was my entry point to my language learning I would say. I’m more interested in taking ielts in English to see how far I went with just tons of passive immersion only.
Under some circumstances n3 can be seen as low b2 isn’t it?
Almost all daily user grammar + 4k words, it it is mastered well I guess it will look like some sore of b2 in conversation? I’ve never taken a cefr test so I probably misinterpreted levels. Anyway I like this chart: https://europass.europa.eu/system/files/2020-05/CEFR%20self-assessment%20grid%20EN.pdf
But shouldn’t it be at least better then b1?
Most resources consider b1 2500-2800 words, and for n3 as far as I know for you need around 3500. Also for me it looks like required n3 grammar is bigger then b1 grammar (question if it covers more constructions you can use)
Am I missing something? Maybe you don’t need to speak in Japanese to pass it and it takes away some sort of skills (listening, better understanding, knowing words you need?), and confidence, so in total b1 is better than bare n3? Bur comprehension should be definitely better as far as I can see
In summer 2025 we’ll have as close to a real answer as possible once they add CEFR estimated equivalencies to JLPT score cards however I’m very confident that N3 will not be considered as a low B2. Having said that I’ve never taken a CEFR test so can’t comment based on experience as a learner. If the descriptions you posted are accurate then N3 isn’t close to B2 at all. This is based on my experience both living in Japan and having met various foreigners with different levels of Japanese.
As far as word counts go, words are counted differently in different languages. For Japanese the average amount of “words” known for an educated adult is around 40-50 thousand. For English I’ve seen this number range from 15 to 40 thousand. This isn’t because Japanese people know more words. It’s because what a “word” is and how to count them is rather ambiguous, especially at higher levels. Basically having a certain number of words in SRS does not tell us much about mastery of those words nor about equivalence to a certain CEFR or JLPT level. Be careful not to fall into that trap.
Regardless, if you wait until near summer 2025 then the organisation that runs the JLPT will publish their research and methodology on comparing JLPT and CEFR and you’ll get your answer. It’ll be a lot more accurate than anything any of us can say now on this forum.
But when you know n3 stuff grasping everything else just from input is not a hard task right?
I’ve seen some interesting construction that I use in English regularly in n1. Right now with only n4 stuff surprisingly a lot of content is understandable! I means I can not understand it, but when I read most constructions are known to me. If I could learn words faster…
Yes, I was comparing it to English native adults. The English test I saw wasn’t trivial for native speakers, it was full of words you pretty much never come across even in flowery novels. That’s why half of the group didn’t know them. However, I don’t rememeber what test it was so I unfortunately can’t contribute much more to this point, my apologies. Given that everyone us mentioning different English exams here it’s likely I didn’t realise how many we have. I just rememeber that the words were extremely unusual to the point where I didn’t think anyone really needed to know them, which is the exact reason I thought it matched with many people’s comments about the N1. It seems I was wrong though, and the N1 has significantly more common words.
Ah, this is not the definition of trivial I was thinking, or I think the above commenter was thinking.
That’s fair and makes sense.
I do think discussions on the difficultly of the JLPT are most important for people who haven’t passed it. When people who have passed it say “No one needs these difficult words” and native English speakers didn’t understand a bunch of the words on the exam I watched (I don’t remember which one it was) it seemed that they would be similar. I was planning to learn all of the content anyway, though, because I did know all of the English in that test. I double majored in English, though, and am cognizant that I don’t represent the majority of the English speaking population.
I think we could all agree on the listening section, but I think the rest of that is opinion based rather than an objective fact, which is probably why there is so much disagreement on what makes it difficult. In fact, this reminds me of how a university math professor might have a disagreement on what makes complex math difficult compared to a student who is not a teacher, if that makes sense. That wouldn’t make the student more wrong, I don’t think. As someone who thought university level computer science classes, this disparity in the perceived cause of difficulty I feel is a common issue. Given that education still hasn’t been solved, none of us fully understands the issue.
I think most everyone here has probably done many, many, mock jlpt exams. It not being free or easy for all people to get to is something you didn’t bring up as an issue that I think many people have as a gatekeeper. Complexities on complexities, but that’s the point of discussion, right. No one’s opinion is 100% correct.
Oh that’s what it was! I was wondering because it seems like you don’t do a lot of output, so I would be interested to see if it ends up like that interview with Kaufman.
How long has it been since you began English output, though, wouldn’t that skew the passive immersion data?
This is great info, I’ve always been curious.
Many people would argue that this depends on the content. Conversations are extremely different from written prose, which has its own voice, as well as can have a narrator. Narrators don’t typically speak like your average citizen. But you could probably find native Japanese people speaking to each other outside a learning situation (or one targeted to English speakers) and see for yourself if you think it would be more difficult, or the same level of difficulty, as the input you’ve been doing so far. The number of reported cases of people saying they felt like they didn’t understand anything the first time they listened to native Japanese speech in Japan seems high enough that I think it is different.
Actually I speak with my sister time to time and practiced 4 basic tenses before switching to immersion method. Also I leave in Italy, which is not helping at all, it’s been 3 years, and when I stared learning Italian 4 months ago I wasn’t even a1, I guess country doesn’t metter as long as you sit at home… anyway at some point I will start speaking to people when I have enough vocab and do little stops, and maybe even take this language to c1 at some point if I want to start business here or something like that. What I mean is that experiment won’t work I with this language in my case.
That will. First of all I was starting from b1 level (or higher, preply test gave me 3400 known words at that time, which is actually even more lemmas as I figured it out). And then I used to to write something on github, write comments under YouTube videos, then I tried writing some technical articles, and I think the endgame was Reddit, but when I got there about a year ago I was writing fluently without using dictionary. As for speaking I stared using English with my family about 3 month ago, and after doing 1 week of only English speaking challenge I gained the ability to start and maintain conversation without delays.
One more interesting thing is that my writing is much worse then speaking, I do a lot of mistakes in text which I wouldn’t say, constantly forgetting to place articles (usually before words with an adjective in front of it) but in speaking I do not make these mistakes as well as some others. Maybe it’s the consequence of those 3-4k YouTube hours I have and really no text except some docs which I go though without reading.
And anyway, most of my constructions are unnatural and I need to relearn them to sound like a native (as far as I know)
When we look at AJATT there are some interesting guys who never learned grammar, for example Can Cannon from last Mat vs Japan’s videos who claims to that he learned Japanese just by watching anime and learning top 1000 words or so. + kids
So I guess I could assume that if a watch and read a tons of stuff I should be able to grasp everything, but more grammar I know initially easier it will be.
The question is how it will be compared to European languages where you need to learn only a few tenses and a bunch of words to understand everything almost instantly
Sorry that I keep using this word but that is fascinating to me as well. If you know enough to speak and read Harry Potter, I would assume passive input would be much easier, but either the bar for that is higher than I thought or you really, really don’t go out . If you’re there, I assumeyou’ll start studying it again in the future? It seems like it should be easier than Japanese will have been.
It’s amazing how fast that works, I was shocked by how much more quickly I can make spoken sentences in such a short amount of time after doing calls and meeting up with visiting friends. I started so slowly that I thought I was annoying my friends, but all of them were very positive about it and the improvement was fast! I’m still not fast enough because my vocabulary isn’t large but I’m so much better.
… Yeah I believe that is physically impossible for me to do . Like, I know there’s a lot of debate over how accurate statements like those are in the first place, but I know my brain doesn’t work that way, even if it is possible.
I’ve been able to ramp up my learning over 5x what I was doing in the last month without eventually losing my ability to learn anything new, but it’s just that I’m at a normal level now. I’ve discovered from the app I mentioned to you before, that I spend a shockingly consistent 210-220 minutes on Japanese (not counting gaming) per day. I want to reduce the time studying and increase immersion, but I’m still playing catchup on how far behind I am compared to people who are competent at memorization for the same number of years.
I can read Harry Potter just because of lingq, and hold meaningful conversation only about learning language, at least for now, with a lot of grammar mistakes no metter i speak fast or slow.
I started doing YouTube after about 120 hours of learning (only 40 of them are reading), also keep I mind that I was about a1 already and could order pizza, so maybe 250 hours in total, a bit less then I spend on Japanese. And now I can watch content about language learning on YouTube quite fluently, and other topics are opening as well.
I really really don’t go out
I guess so, I’ll settle here for at least next 4-5 years. I gues I’ll start doing 3h day after February next year, it’s not clear data because I’m not starting from scratch, therefore it’s not that interesting, also I have no idea how to use it, even I’m my country I set at home all the time
Which app is i?, I missed or didn’t pay attention, probably
Oh, memorization of the words is so bad for me!
Retention is terrible, no template changes work, kanji were like so much easier, I could do like +80 in 3.5h ! Here I can not even add 15/h and retention is terrible! (Idk if I can add 10/h, so bad it is). And what are you staging with?
Just the script I wrote that tracks minutes I spend doing dedicated Japanese study by watching which applications are open.
I am sure I couldn’t do 15 an hour. I switched to practicing words immediately upon seeing them and decided to add them and I also… Probably am going to go back to using Anki so I can take advantage of FSRS 4.5.
Oh ye, I remember, just didn’t get for the first time that you were talking about tracking itself.
How many words do you think you know already?
I’m adding words that I encountered at least 2 times, or I know that it will be used not once during the week of input
I wonder how they count words. For example, are 協力 and 協力的 counted as the same word? Does this take into account the different ways that a word can appear or does it just count the base word? I’ve never been able to find. Concrete answer to this question.
The same goes for English words. Would “possible” and “possibly” be considered the same word or different words?
I find it interesting.
) but that doesn’t mean I think other people should do things differently; they aren’t me.
