Feel Like Immersion Isn’t Working for you?
A tale of two methods: Anki and Immersion. The two pillars of the learning method promoted by this website, yet almost complete opposites. While Anki excels at improving one’s reading, listening immersion is the essential method to improve one’s listening. Where Anki provides one with a precise amount of daily studying that must be done without fail, immersion is almost structureless: watch something, add the audio to your mp3 player, listen whenever you can.
Where progress in Anki is easy to assess (card count, correct answer stats, etc…), progress with listening immersion is extremely hard to gauge, if not actually impossible to do so at early levels. And while Japanese learning websites feature a wealth of extremely detailed advice on how to get the best out of Anki, with distinct phases (RTK => J-E => J-J) and almost step by step advice for each, advice on listening immersion tends to be restricted to general guidelines at best.
And while some have come to dread the demands made by Anki , others such as myself have come to appreciate it for the clear sense of purpose and reliable progress it provides, while at the same time scorning listening immersion for its seeming inefficiency. There’s no better description of this latter mindset than the following excerpt from a recent comment posted on the website by user Vi:
I have no problem with spending huge amounts of time studying Japanese, but I DO have a problem with doing it without direction, without a plan. Before I started using Anki, I spent years playfully toying with Japanese as a hobby. I learned a few words, the basic grammar, and delighted in making out the occasional word while watching anime, but given a Japanese book, I couldn’t tell you what it says. Thanks to Anki and sites like this, ~2300 kanji cards(including ~1300 J-J sentence) and 3000 more J-J sentence cards later, now I can read that book. Thanks to the order and purpose provided by Anki. Order and purpose that I’m now to chuck out the window in exchange for blind faith that if I spend enough months listening to Japanese from the moment I wake to the moment I sleep, that I’ll be able to make it out.
These words resonate deeply within me, as I’d have expressed essentially the same opinion even a few months ago. Yet here I am trying to convince the unconvinced that listening immersion works after all. How did this happen?
I used to suck really, really badly at listening immersion.
I should probably start by pointing out that reading is my main goal in learning Japanese, and so when I started on my journey, I viewed listening as just a perk that I’d either pick up “on the side”, or possibly by putting a little extra effort after my reading got to a level I was happy with.
So when I first found this website it was immediately clear who my number one trusty sidekick would be: Anki. But the whole immersion business sounded pretty important too, and on top of that easy too: just have Japanese material playing in the background whenever you can.
It all started well enough. I tracked down a Youtube playthrough of my favorite JRPG, ゼノブレイド, downloaded it, erased everything but the cutscenes, and added the audio to my mp3 player. 11.3 hours of something I love, dense in speech, and with varied voice types. Looking back I don’t think I could have chosen a more perfect first material to add. So I made sure to listen to it frequently, and made some effort at finding online Japanese stations I liked to listen to for a little variety. Not the most amazing start at immersion ever, but certainly a solid first attempt.
>>Fast forward 12-13 months>>
One year later and my Anki progress was kicking ass. 2000+ J-J cards in, well past the hard transition into J-J, and when testing that knowledge against native media the milestones kept coming. For instance, I’d just finished reading my first novel (extensively), and had just for the first time fully understood a 15 page text heavy furiganaless manga while only having to look up a mere handful of words.
“What about immersion?” Well, my mp3 player had this 11.3 hour long walkthrough of this RPG I love and… nothing else. How did this happen? Now, I don’t usually talk to my mp3 player, but if I did, here’s what our typical monthly conversation during that year would have looked like:
Yeah! We should totally hang out more. And I should get you more content too.
What? How about tonight, you say? Oh, sorry. I’ve booked this week with Anki. We’re really close to a breakthrough, you see?
How about next month?
Again and again I neglected immersion because why waste precious Japanese time listening to nearly incomprehensible noise when I could be doing something I knew worked?
Excluding the sound in my Anki cards (mostly text-to-speech, with a few exceptions), my full listening experience consisted of nothing but sporadic usage of the immersion mp3 player (more because I felt “I really should use the thing once in a while” than because it seemed to work) and a serious attempt to watch One Piece (which I never watched in english) without subtitles.
The latter occurred around my 9th month, and while it did feel better than previous listening experiences, I could only take so much of “understand basic words here and there and maybe an important one every five minutes”, and so that too I dropped at the end of Season 1.
So how did this sad story take a turn?
I started sucking a lot less at listening immersion, and even came to like it.
Looking back, two main reasons made me feel I shouldn’t keep pushing listening immersion away much longer.
- At 2000 J-J cards, it becomes clear that the hardest stages of learning how to read have been cleared. While the remaining path is still long, it’s also a cakewalk compared to the fiendish nature of the first 1000 J-J cards.
Japanese is filled with words expressing emotions, and reading these is no replacement to knowing how they sound. A good example is 一体(いったい). Checking it’s definition on, say, dic.yahoo.com (強い疑問や、とがめる意を表す) can’t complete with hearing even a single “一体どういうこと？” in context.
Even so, I was weary of disrupting my efficient hard earned Anki routine, and felt I needed some jolt to move me on, so I went and bought a session of this site’s Personal Advisor essentially so I could ask “is it really worth it for me to sacrifice some of my limited Anki time for the sake of sneaking in some listening?”
The answer I got on that point was a resounding “yes”, so I implemented that advice. Nothing drastic. Definitely no “ listen every waking hour!” I merely:
Reduced my new daily Anki cards by around 20%, and waited a little for the reviews to subside, buying me the daily 25-30 min it takes to watch an anime episode, which was then added to my mp3 player.
Started being more aware of turning the mp3 player on whenever I had spare time, or was doing a menial task.
After around two months of this, a turning point happened. While biking to my university with ゼノブレイド on in the background, I suddenly noticed that I could make out the spoken syllables apart with perfect clarity. Not the meaning of the words themselves yet, but this was a big breakthrough.
Somehow this simple fact changed everything. Immediately my motivation to bother with passive listening skyrocketed, and thanks to the substantial vocabulary I’d already built via reading, I started consistently recognizing something new with every single listening session. The way this tends to work is that after you recognize a word it’s much easier to recognize other words in its vicinity that you already know.
What’s the moral I think should be from taken from my experience?
Listening immersion does seem to eventually work if you at the very least do some (not zero) of it, even if you don’t really do it well. And if you are in a situation similar to the one I was in (having a strong vocabulary but not even able to make out the sounds), your problem is actually pretty specific. The moment you do become able to make the sounds out clearly, the proverbial floodgates are open.
But how to “open those floodgates” then? Certainly the normal immersion method is what did it for me, but given such a specific problem one might expect a specific targeted solution would be preferable. While thinking about this I realized something I have been using for the past few months seems ideal for the task.
A method for “making out” the sounds
Who is this for: anyone who reached a high enough level that they can make Anki cards (so being friendly with Anki is a requirement) of most sentences they find with some ease, but when listening, have trouble recognizing the sounds.
Main tool: The Subs2srs tool (Follow the link for an explanation of the tool. I really like the settings Cayenne uses, so I recommend those too).
Take some long series you like (anime, drama) and fully transform the first few episodes into anki cards (with sound from the original series). Keep these in a deck separate form your main deck.
Go through the cards in order, deleting the ones with no new words/grammar (there should still be a healthy quantity of these). Keep the other cards and add definitions to them. Move those to your main deck.
Rip the audio from those episodes you turned into Anki cards, adding them to your immersion ipod/mp3 player.
As you start reviewing those cards in Anki, make sure to also find time to listen to your newly added material.
- After doing this for enough episodes, try watching (and also continue adding to your mp3 player) the remaining episodes without turning them into Anki cards.
This Anki phase creates a listening environment where correctly identifying sounds is the main challenge, hence you train this skill much faster.
After giving the Subs2srs treatment to the first few episodes you should be very attuned to the specific speech patterns of the characters in that series, so when you then watch the remaining episodes, you can focus on recognizing those words you know.
You get a fast source of excellent Anki cards. I feel that cards obtained this way aren’t just better looking than my normal ones, they’re even a time saver (which is why 12 of 19 of my current daily cards come from this tool)! Talk about a great deal . . .
There is always a lot of talk about immersion on this site in posts and comments. I’m hoping that this post addition helps some people fit in the final piece of the immersion puzzle.
Anyone else struggle with the concept of immersion and make a breakthrough? How did you do it? What do you recommend to those still in the heat of battle?
Written by: Alexandre
Mathematician & Immersion Master. Analyzing the formula behind Japanese greatness.
I just started using subs2srs last week and I love it it’s a great tool. I can already see results. This was a Greater read my friend. My problem is the opposite lol I listen and do anki but I have trouble reading and making the switch to j-j sentences :/
Glad you enjoyed it!
And making the J-J switch is definitely very challenging. In fact, though this post is mostly about the difficulties I had finding the motivation to do listening immersion, I’d still consider the J-J switch as the biggest hurdle. I can’t even really say that I adquately understand what are the key ingredients necessary to succeding in that transition, but what I felt I was doing was this:
1) Track down something I want to learn (generally words from a manga)
2) Do some mild amount of branching on the dictionary starting from that word, capping the branching after a few (4-5) iterations using english if necessary (and maybe I should point out that in plenty of these cards the understanding I had when creating them was definitely hazy).
3) Repeat the above
It definitely feels like a struggle in the beggining: a scavenge for the little things you can understand in a sea of things you can’t, which are then immediately enshrined in Anki cards lest they disappear forever.
But eventually somewhere in the 700-1000 card range (Adshap mentioned this elsewhere, and it matched my experience, at least) things just start clicking, with usable sentences starting to feel like they really are everywhere.
If I were going through J-J again knowing what I know now I’d try the following changes:
– Start the branching process in step 1 above with Subs2srs cards
– This website’s “Branch Annihilator”, which wasn’t available when I went through J-J.
Thanks for sharing your perspective and advice. Now I understand why there are people so into anki, not just because they believe it works and see progress, but also because they can measure it. I’m realizing more and more that people have different passions, motivations and skill sets, and that’s perfectly fine and nothing to argue about. Such as monoloinguists who stick to learning one and maybe a side language (like me) vs polyglots who love learning multiple languages and can’t settle for just one. And in this case, people who lean towards Anki and people who lean towards immersion. I wonder if it has anything to do with left brain right brain types. I’m very right brained. But I can see someone being very logical using anki. Or perhaps it’s more personality wise.
Great insight into the importance of measurements. I don’t think I’d actually thought about the issue in those exact terms before.
This is a bit far afield, but the creation of better standardized measurement systems was one of the crucial steps that allowed science to boom way back in the Renaissance. Being able to measure things precisely simply allows for completely different levels of control over the phenomena being measured.
As for the rest, I’m definitely a “logic first” kind of guy, so I most certainly fit into what you are saying, but there’s seems to me to be a crucial difference between the monolinguist/polyglot and the Anki/immersion issues. Namely, the first seems closer to be measured binarily, as you’re pretty much either a monolinguist or a polyglot depending on how many languages you speak, while the second issue is probably better measured on a spectrum, with me being on the higher end of the “Anki” side, but many people being in a middle ground.
That’s very true. I agree that Anki/immersion use is more of a spectrum.
I’ve had people from the poly-linguist perspective tell me, “Oh, you’ll get back into Chinese later,” as if it’s a language to “collect” and then likewise have seen people who focus on one language say put down polyglots (or aspiring) by saying they’ll never really get good at the languages having their attention diverted or aren’t spending enough time with one culture to really understand it. There are people who just have no passion for learning more than one language. And there are people who don’t want to be kept down by one language. So it’s just another one of those debates like Anki/immersion. However, Anki/immersion is definitely more complex than learning multiple languages or not, because it’s not just one or the other, but as you said a spectrum.
Yeah, I knew what were getting at. My point was mainly that since in the Anki/immersion thing there is an actual “middle ground” to speak of the conflict is less intrinsic.
Off topic, the monolinguist/polyglot issue is pretty interesting.
Personally and at the moment I’m definitely on the monolinguist side: there simply is no other language out there for which I would possibly have the same kind of solid motivation to learn (unless you counted english, of course), and the abstract notion of “picking up languages” doesn’t seem that appealing to me.
Yet it’s not that far-fetched to imagine me moving to the other side in a few years once my Japanese is solid. My reasoning here being that having learned a language as out there as Japanese will have taught me so much about how to learn a language that even a milder interest in some language will be enough motivation.
It’s shocking though, to start all the way at the beginning again! But yeah, I’m with you. There’s just no other language that I’m so passionate about. However, I am learning a second, that being Japanese Sign Language, but it’s within the same country and is of a similar culture, so it’s a bit different than just picking up a new language from another country. I was learning Mandarin, but my passion just didn’t hold up to the extent that it did for Japanese, so I cut it off as a distraction. I’ve heard of polyglots that are always learning new languages every year and I think, “Ooh, that sounds cool.” But it’s not a passion of mine. It just sounds interesting from an outsider’s perspective on what that must be like. I don’t imagine myself learning any other languages aside from Japanese and JSL, unless something happened in my life in which I needed to learn another language (sudden urgent move to another country?). Then at least I’d have the skills to learn another language.
But it’s not really like starting from the beginning again, though…
First because you keep all the language learning techniques you perfected. Maybe a good video game analogy for this would be to restart a Pokémon quest with the same team you previously used to beat the Elite Four now rescaled to level 5, but keeping the same move pools. Their level might say “puny”, but their attacks would be devastating.
Second because some language quests are just more accessible than others. Around my second year in grad school I made an half hearted attempt at picking up Italian because I have far too many Italian friends who can’t seem to stop themselves from speaking in Italian when together. My first (and only, really) move was to ask one of them for some book in Italian for me to read (a translation of “La casa de los espíritus” was what I got). I actually read the book just fine (my native language being Portuguese), but didn’t really feel the proper motivation to continue to other books.
And it’s really striking to compare that experience to my Japanese learning experience. I mean, in Japanese my move into text heavy media is still recent, required more than a year of intense study prior to it, and I’m not even sure whether I’ve read something at quite the equivalent level of “La casa de los espíritus” yet. But I do know that when I do manage to it will feel like a great achievement, while the Italian experience was very much “Meh, of course I can read that”.
I’ve tried Subs2SRS on several occasions for a few different dramas. I really like the idea of it. But I found that I could never make all the cards in an episode sync properly (the timing of the video clip would always end up off from the line, etc), no matter how much I fiddled around. Do you ever run into this?
It’s not clear to me from your post what your problem is exactly, so let me list the ones I can think of.
1) The timings of the audio and the subtitles gradually drift away as the episode goes on. I.e. you have the first line synced perfectly, but as you advance it starts getting off in a progressive way.
2) The lines cut of too soon. I.e., even when your the audio starts precisely at the beginning of the line, it’s too short and you don’t get to hear the end.
3) The timing of the audio matches the subtitles just fine in some big chunk of the media, but it’s off elsewhere.
1) This one never happened to me, and it might be hopeless. If this is the case it might be better to try your luck elsewhere. I currently have over 800 active (i.e. already being reviewed) cards from Naruto, Naruto SD, Code Geass, Bleach and Ohitorisama in my deck, so there definitely are good subtitles out there.
2) By fiddling with the “Pad Timings” option in Subs2srs I can generally fix this for all but a minority of cards, which I generally have to delete, but that’s not that big an issue.
3) This I do find a lot. All the instances I’ve seen of this seem to be caused by the intro segments or what I guess are pauses for commercials, which never seem to last the same time in the audio as in the subtitles.
What this means is just that you need to run Subs2srs once for each of the chunks of the episode with the subtitles shifted differently for each chunk. The main wrinkle here is that I find the “Time Shift” feature in Subs2srs insufficient, since the cap in the allowed shifts is inferior to the shifts I often need. The solution I use for this is to shift those timings directly in the subtitle file using Aegisub.
Hope that helps.
Hi! Thank you for the in-depth response! Sorry that I didn’t explain clearly.. But yes mostly what you mention in number 1+2. I’ve tried fiddling with the pad-timings without much avail, as it just ends up fixing one part of it, but will progressively get off-track again. :/
Maybe I’ve just had bad luck with the dramas I have tried.
Well that certainly sucks.
But the good news is that when you find one that works you should be all set for a while.
Good luck with your next picks.
Hi Alexandre, thank you for your input.
There is something I don’t quite understand. You say “[…] I started consistently recognizing something new with every single listening session.”, and this seems to be what you define as “turning point” and “breakthrough”. But this is just listening the same content with repeats. Can you understand better OTHER material that you watch for the first time? Have you tried with various genres to get a more representative sample of audio media and be certain that your listening comprehension has indeed improved?
(e.g. if you Subs2SRS Gintama, which is a HARD anime, you’ll certainly end up understanding most of it; watching a high school romance anime just afterwards would make it seem you understand a lot more, but it would actually lie in the relative difficulty of the contents rather than in an actual improvement).
Also, if you (and certainly many other learners) enjoy the structured and trackable learning with Anki, why not simply use audio sentences? The Core deck contains 10k of those. I’m pretty sure if you SRS all of them, you won’t ever have any actual LISTENING problem. However, a vocabulary problem, quite possibly.
First let me address your main question: has my understanding of material I watch for the first time also improved?
The answer is yes.
The whole “trying different genres” issue doesn’t really apply very well because the “breakthrough” was only about 2 to 3 months ago, which doesn’t really amount to a representative listening experience (though there is a modicum of diversity in it) since my focus is still firmly on reading (and I’ve had some incredibly satisfying recent breakthroughs on that front that gobbled up most of my japanese time).
There’s also I think a further implicit question in your comment, which is something like “why would being able to listen to something better over repeated listening deserve to be called a “breakthrough”?”. A few points on this:
– Passive listening is extremely important because it allows you to turn otherwise dead time (like time you spend going to the supermarket, cooking, jogging, etc…) into a studying opportunity.
– Passive listening does A LOT MORE than just improving listening comprehension. In fact, some of the moments I found more satisfying about it were those where I figured out a new word from the audio (complete with the kanji), or when you hear a word in a context which is more broad than the mental definition you had for it. And on top of all that, passive listening reinforces your reading (and vice-versa).
So given this, I think it’s fair to say that passive listening is too good a tool to pass on, which is why the moment I finally felt I could make it work deserves the term “breakthrough”.
I’m confused as to your suggestion, though. Namely, what do you mean by “why not simply use audio sentences”? Are the “audio sentences” you are mentioning fundamentally different from the cards that would be obtained via Subs2srs?
I also let aside my immersion but in my case I think it’s more because I can’t seem to stick to it. For example, on my ipod I put anime audios of a few different anime of different genres that I loved (death note, persona 4, air, to name a few), I also have about 100-120 songs only in japanese, and I am even subscribed to 5-6 podcasts that update either once a day or once a week. But after a week or two, I always end up shifting back to the english/other language stuff. I can’t seem to find a balance, is what I’m trying to get at, I think. I don’t know if I just have the wrong material, or if this is common for someone in the RTK phase, or whatever, but man I don’t know what is wrong.
What about trying to make listening to your immersion MP3 as habitual as possible? For example, put your MP3 player by your bed when you go to sleep and when you wake up in the morning the first thing you do is pop in the headphones and listen while you make breakfast, brush your teeth etc.
If you drive, it could be helpful to leave some CDs or plug-in your ipod there. A few years ago, I was torn between talk radio and Japanese. But one day I just made a rule that I would only listen to Japanese in my car. So it’s a guarantee, unless I don’t go out (which rarely happens), I’ll always be listening to Japanese.
well I take the bus but that might be even better since it takes longer haha, I guess I could try to force myself until it becomes a habbit
On the bus you can actually watch the video too. That’s nice. You can’t do that while driving, ha. I like to read manga on the bus. I’m only on a shuttle bus for a short while from the parking lot to my school.
Ohh well I can’t exactly watch while on the bus as I do my anki reviews there, but still I get what you mean haha. Although that does sound like a great idea for coming back from work… I’m in my 2nd bus for ~30 minutes so I could watch an anime :o
Alexandre you are a genius for that implementation method. Can’t think of a better way too learn Japanese than through the anime dialogue + anki; going over the cards and then listening to the lines afterword. One episode is a great deal of work sure, but if you use – not just anime- but anything that has some good to great dialogue, you’re talking learning grammar and vocabulary at a relatively faster speed than just using your college textbook or whatever have you. This could easily turn at least half of someone’s day into Japanese study!