Why Japanese-English (J-E) Is Better Than English-Japanese (E-J)
Notice how all methods discussed on this site always refer to J-E but never the opposite E-J? You look at a Japanese sentence, see how well you know it, then check the English to see if you understood it right.
But what about the opposite? Seeing an English sentence and having to produce the Japanese. Afterall, wouldn’t this practice your production ability?
There are multiple reasons why this is done, but it really comes down to one concept: Your goal is to understand Japanese.
You will eventually be reading Japanese. Watching Japanese. Listening to Japanese. Not English. So doing the opposite would be unnatural.
But what about the English side!? It’s still there.
Recently Jalup reader Alexandre made a great comment helping someone who was concerned whether she had to understand every word in the definition section (back) of an Anki sentence card to mark it as correct. He responded:
“Failing cards for not understanding something in the back of the card is something you should not do. In fact, if you manage to understand the sentence in the first place, there’s no reason to even look at the back unless you happen to be unsure about something, in which case you would look at the definitions in the back merely to see if they confirm your understanding or not.”
This is what it all comes down to.
The goal is not to do J-E or even J-J. The ultimate goal is to do just J.
You read a sentence on an Anki card, you understand it, you don’t even look at the back of the card (the definition section), and you move on. This is how your interaction with real Japanese will become, and this is what is leading you there.
Through this movement, you go through reviews quickly and efficiently. And this is why seeing English on the front of the card would get in the way.
Are you saying E-J is bad?
You don’t need me to remind you again that this Japanese study is your own. Do you like doing E-J and have been successful with it? Then keep on doing that. Different person, different strategy.
But what about RTK or Kanji Kingdom? The English is on the front!
Some of you may point out that this is E-J in a sense. If I’m so adamant about J-E, why not go kanji (which is J) to English keyword.
Two important reasons:
1. The English keyword is more of a mnemonic to create an association with the kanji.
2. You want to be able to practice producing kanji. You can’t test yourself on this if you see the kanji first.
I’m aware though that the later issue of still being exposed to English is present. To cancel out this effect, I include Japanese keywords. Kanji Assist solves this a step further by allowing you full access to the Japanese keyword immediately. Within a short amount of time, you are really just looking at the J-keyword and then producing the kanji. Similar to the way a native Japanese student might do on a test in Elementary school where he has to write the kanji from the hiragana word.
Take the J-path. It’s the one you want.
Short term, long term. The benefits are good. And you want good.
Founder of Jalup. iOS Software Engineer. Former attorney, translator, and interpreter. Still watching 月曜から夜ふかし weekly since 2013.
Great article! I really hate looking at the definitions for my J-E cards but I pretty much always read the definition on my J-J cards, even when I 100% understand a word. I guess I just like the reading practice, even though it probably slows down my reviews a bit. Plus it feels like I’m marking myself correct, which always feels nice :P
On the flip side (literally), the front sentence on the cards I’m not as lenient with. However, I do mark cards correct if I see a word like 検討 that I know the reading of (and remember that I understood it when it was learnt and/or reviewed) but not the definition, only if the sentence still makes sense overall. That probably goes against the reviewing instructions ‘Choose 1 if you messed up on the pronunciation or didn’t understand a WORD or the full sentence’, though like Alexandre has said previously, given the nature of JALUP decks, I know that those words are going to sort themselves out because they the deck is built on itself. It’s definitely something that I just realised by doing, and concerned me a lot at the start of J-J especially. Though I’m not entirely sure if my approach to unknown words -in example sentences- will change when I am making my own cards and don’t have the peace of mind from a pre-made deck to assure me that’ll smooth itself out. Otherwise I feel I might spend an eternity branching every single sentence unknown until death, which didn’t work my first time around. But, maybe I won’t have that problem after the intermediate J-J deck! Or if Adam releases more decks :P *hint hint*
As an aside, I really think Alexandre should write some articles for this site if he’s interested, or at least a place where his thoughts and ideas can be gathered. I’ve gained much insight just from reading his comments around the site, and he’s quite well spoken and has a lot of useful knowledge to share. On top of that, he’s also one of the few vocal people on here that have actually done the 9-10,000 cards that we’re all striving towards. I can only imagine he’d have shared many of the perils and pitfalls that all Anki learners go through/will go through, so I’m certain we’d all relate and benefit.
True. If it’s Japanese, reading practice is reading practice. Enjoy it now. But later on, when your cards are in the several thousands, and you have that super exciting manga you are just itching to blast through, you may have the urge to go a little bit faster.
Alexandre is a veteran, has been on the site for a long time, and is a great writer. He actually had written one guest article (http://japaneselevelup.com/listening-immersion-doesnt-work-for-you-or-does-it/), but I’d always welcome him back for more.
I guess it’s my mild self-diagnosed OCD versus my Japanese leisure time, I think the latter will win out eventually!
Yes! His article on immersion is great! Totally inspired me to start immersing much much more a couple of months ago (in collusion with your own words of course)! It’s nice to see someone (besides yourself) who have actually gone through the JALUP journey at its core. I’ve actually been reading the site for a little over a year now and he’s one of the few writers on here whose words have a weight close to your own (shirobon, Rachel M and Eric are a few others). Some of the best info on this site is scattered throughout the comment sections that have been buried over time! It’s fun to find some of the older stuff when random articles pop up though
Since it’s been such a long time since his last piece, I’m sure he has even more he could share! I would be very interested in any advice he can give towards the J-J process, and especially in regards to Anki and creating and reviewing process. His words already have changed some of my attitude towards study, I can only imagine that’s a drop in the barrel of his 2-3 years of experience with Anki. With the intermediate J-J deck I feel more prepared than ever, but I know there are still lots of neat tidbits, tips and tricks that could help smooth out the process even more so(like his advice on Subs2srs, his branching cap and the value of immersion)! A separate pair of eyes on the same topic can be quite useful in my experience, especially since his JALUP path is even closer to site users than your own!
It’s quite nice to hear that you found some of my comments useful.
I quite like contributing to this website, partially because I’m just the kind of person who loves explaining stuff to an interested audience, and partially because I’m quite grateful to this website for the role it played in allowing me to learn Japanese in a much more effective way than “usual”.
The catch, of course, is that I don’t really have that much free time these days. In fact, shortly after my article about my troubles with immersion and how I’d finally come to grips with them, I was thinking about writing an article about what you might call “the puzzle analogy”, which I felt was a particularly good analogy for what I had been experiencing at the time (somewhere in the 3000-5000 J-J card range). However, I couldn’t really put it together before the end of the summer, which was when I started teaching for my very first time (having finished my PhD last year), which is a handful, so that article never came to be. In fact, this last semester I was so busy I sort of took a “hiatus” from commenting here until I finally hit the 10000 J-J cards, because I didn’t feel I had the time to spare for my standardly long comments.
As for card making tips, Subs2srs is a big one, as is trying to make some effort to have some of your reading immersion come from stuff that allows copy paste such as websites. I can’t think of much else right now which isn’t somewhat specific to me and my personal tastes.
But really, I doubt you’ll find it that big of a problem. The transition from J-E to J-J really is the hardest hurdle in the method by far, so much so that there really isn’t much in it that compares, and those first 1000 J-J cards really seem to be the key, so Adshap really has you guys covered these days!
Once past that adding new cards really feels a lot like “more of the same”, clearing your vast ignorance slowly but surely.
It’s even hard to think of proper hurdles past that point. The only one that truly comes to mind is “finding your first novel”, since there’s no good way of knowing when you are finally at a level to read something you enjoy other than trial and error.
That’s completely understandable, life gets in the way sometimes! I don’t know whether to be disappointed or excited that there are no real stop gaps (besides finding media) that’ll hindrance my future studies. Sort of sad that the challenge will be waning, but also fairly liberating and no real reason to ever stop now.
I guess my only concern is card creation, mostly because it’s something I don’t like doing. Especially because I always feel like any card I create isn’t going to be as good as someone else’s, or there’s more relevant vocabulary to learn right now. That kind of goes away with the premade decks of adshap, plus having to make your own deck takes me ages! My first branch was like 100+ cards and I know that’ll be greatly reduced after 1000 intermediate deck but it’s still a burden on my limited study time.
I’m so glad adam has decided to make the advanced 1000 though, it’s only going to make the card creation process even less daunting when I eventually start it. I’d love for these premades to get me to a point where I can just use the One deck and introduce cards like any other premade, no branching required. Sounds a bit like a dream but it would make learning new japanese an absolute breeze.
But even now, you telling us that there aren’t really any more big hurdles to jump over is advice in itself. A pretty reassuring tidbit I have to say. Though I guess the ‘unknown’ part of my future study isn’t as ‘unknown’ as diving into J-J for the first time.
In my experience learning English, using Anki to train production isn’t effective. What I noticed is that you can always answer a card correctly, but never remember the damn word in real life when you are speaking with a native.
For me, what made using words easy was to frequently write. It is different from “anking” because all you mindset is in the language you are writing and you are training your brain to produce the language.
It is not enough to became a good speaker, but you will practice not only to recall the words, but to produce entire sentences and arguments in the language.
This is the recall vs recognition argument. I’ve found if I can recall (produce) a word, I can easily recognize it in context. But just because I can recognize a word doesn’t mean I can produce it (best example for how this memory thing works: kanji. I can read waaaay more kanji than I can hand-write). For that reason I’ve been practicing recall.
…but what you’re saying makes sense too. This is learning specific to language. Recognition is easier, but at the same time deeper, because you see a Japanese word and while it may have an English equivalent (cat – 猫, pretty straightforward), often times the definition will be more nuanced. You don’t want to be thinking from a translation mindset either. There’s been several instances where I’ve been reading a Japanese book or something and read a word just fine, and then go to study anki and that same word shows up and I fail it because I thought of the wrong word based on the English definition. Clearly I understand the word, but do I understand it well enough to produce it? I never know what to do in those situations; usually I just fail the card.
I switched my anki format over to J-E after reading this article. I’ll try it for a week and see how it goes. I still feel like I’m cheating doing this, like it’ll be too easy, but hopefully it’ll get me away from English. If it speeds up my anki times too, I think I can live with that ;) I’m primarily learning Japanese so I can read things, and this format directly practices that skill so I don’t know why I’m fighting it. Years of being taught wrong in school, I guess. Here’s what I’ve got:
Japanese example sentence (with word)
Japanese word with reading
Japanese example sentence with furigana, and Heisig keyword shows if I mouseover kanji
Small text: English definition
Small text: English translation of sentence
The back has all the bonus info I could ever need, so I save time looking stuff up. I rarely use that Heisig kanji option but sometimes there’s a rare kanji or I do a little mix-up, and it’s great to have the keyword right there. Hopefully I won’t ever look at the English part, but it’s there if I need to double-check my understanding.
Does that seem like a good set-up?
I’m tempted to add more Japanese example sentences (I just have one right now), though I don’t want to increase the amount of time I spend reading too much… I guess for words with multiple meanings, I might add more sentences.
So, what about cloze deletion? Aren’t MCCD super popular? I know that’s a AJATT thing, maybe you disagree, because it is based on production. Would you recommend using cloze-deleted sentences and having to produce the word, if the card is J-J?
That theoretical situation is to ask, are you saying go J-E to avoid English despite that recall is better than recognition, or even in J-J translation-free learning, would you suggest practicing recognition over recall?
I’m sure it depends on what people are trying to accomplish, and everyone should choose the method that works best for them. If they just want to read, recognition over recall. If they want to be an interpreter, recall over recognition. But for the sake for argument, let’s just pretend we’re trying to find the most Perfect, Complete way to learn vocabulary in general. I’m curious how memory works here and I want to discuss that, but no matter how this conversation goes I’ll format my anki cards however they work best for me.
This is kind of a long-winded and winding comment. But I guess I’m not totally sold on the J-E thing. Any thoughts anyone has on what I’ve brought up would be appreciated. Thanks!
For your card setup, you may want to consider making it even simpler than that. The format advocated for on JALUP looks like this-
Front: Japanese Sentence (when doing reviews, you do not necessarily know which word the card is “testing”)
Back: Japanese Sentence w/ furigana
Back: English explanation of the new word/grammar ONLY
This forces you to exercise your understanding on the entire sentence, rather than focusing on just one word. It’s actually a pretty fun approach, and has some interesting side-effects. You quickly adjust to this process of evaluating and “solving” a full sentence. Before long you get so good at it, that your brain begins to grasp most or all of a sentence BEFORE you actively translate it to English. At times I’ve found myself cracking up at a funny sentence instantly, as I’m reading it. It’s really cool – and never happened in 5 years of Recall-focused Spanish study.
So at this point, a J-E card goes like this for me-
#1: Look at the sentence.
#2a: If possible, read the sentence out loud, as if I was acting it in a play, with appropriate emotion attached based on the meaning of the sentence. For example, a card could say something like “大丈夫？何をすればいい？”, and I’d read it aloud in a concerned, empathetic tone.
#2b: If I don’t understand the meaning well enough to do 2a, stop and mentally translate to English. Then do 2a a couple of times.
#3a: If I’m able to do 2a (at any point before clicking Show Answer), mark Good/Easy and move on.
#3b: If I missed something (Kanji reading, word meaning, etc), mark Hard/Again and try a bit later after having practiced 2a a bit.
For most cards I don’t even bother looking at the English anymore unless there’s something I feel unsure about. I’ll just read & understand it in Japanese, and immediately move on. That’s the beauty of this setup – as long as you understand the sentence, the definition is irrelevant. If you can read it and it makes sense, you’re good to go.
Where does that leave the Recall vs Recognition question? I really like the explanation provided in this article-
Anyway, good luck trying out your new format. I hope it’s helpful for you :)
Thanks for your helpful and quick reply, Matt!
I changed my cards to your recommend format. Suddenly my vocab cards don’t look like vocab cards anymore. They’re just…sentences, that I read and understand. It feels really weird not having a specific word to get right or wrong.
And you know what? Already this feels more like “time with words” than “homework.” Flashcards are boring. But reading? I love reading. And sure, this is reading the same sentence over and over (every few days or weeks or years, or however the cards are paced), but at least it’s pretty easy and quick reading, unlike the frustration some books bring. So I like it more than the old format.
Actually, I suddenly want to start sentence mining again… I’ve had some trouble sticking to anki in the long-term before (in the past year, there’s been 3 long periods when I stopped reviewing) but I think this will really help me. I still feel a little guilty, like it’s less perfect practice (since there’s no production part), but hey, if this allows me to finish my cards in half the time, I can take that extra time and skype a Japanese friend and practice speaking…right?
Thank you again Matt!
And if anyone else has input, I’d still love to hear it!
Oh! I wanted to mention too, I like your idea of “acting” out the lines. I usually say them out loud, but not with emotion. I’ve been in Japan for the past 2 months and something I’ve noticed is all the “funny voices” people use when speaking Japanese, but I guess that’s really just emotion (emotion, and a little exaggeration because they’re telling a story and want it to be entertaining). My roommate always says どんどんどんどん like it’s a song or something, you can feel the steadily building ness from just how she says it, where I recite it from a textbook. I think more rhythm and emotion in how I pronounce words would be good for me. Plus, it sounds like fun! So thanks again, Matt!
Awesome! I’m glad you were able to shake things up a bit and make your reviews more fun in the process, and happy that I could be of some help :)
I definitely have that feeling as well, that it’s just fun to read. It doesn’t feel like work at all. Now if I could just find a way to enjoy my kanji study time to the same degree, I’d be in great shape >_> heh
You worry too much about production. I don’t think Adam expects any of us to be able to produce the words we introduced from Anki cards by themselves. It’s Anki and the combination of immersion in native Japanese materials and people that allow you to truly encapsulate the meaning and production of your vocabulary.
Also, if you are worried about your card creation I highly recommend the beginner vocabulary deck from this site. It takes out the worries creating the best foundation for yourself and you can spend more time learning rather than choosing what to learn when you don’t really know anything yourself. It does cost money and I know that can be an issue for some people. But if you have the means to supply yourself with this invaluable tool, I urge you to do so!
You know, another thing that’s cool about this is that mistakes I would make reading a book, so I can kind of filter those out while the correct version is right there and fix them early on. If I was reading a book I wouldn’t even notice I was wrong until later, when things stopped making sense. For example, I just mixed up 親しい with 新しい. I hadn’t even noticed how similar they looked until just now…
When I start J-J in about 500 cards here, I’m planning to buy some help from JALUP (that PDF and if I have trouble after, maybe a deck from Stage 2). I’m sure it’s all great stuff. But if I can get a community-built core2k deck for free, or get just half of that deck for $100 from JALUP…ehhh I’m just a poor college student.