More immersion or study for vocabulary?

Japanese
#1

Hey,

My question is not a general one, I would like to get your opinion about my situation. I have been studying Japanese for a while, I mainly use Anki for my vocabulary, I have a deck with around 8.000 words in it and I finished around 5.000. So on paper I have around 5.000 words of vocabulary. I am at last part of the deck but it takes time because words got harder (and similar to each other).

People always say “immersion” but my immersion has been mostly about Anime, Japanese games and rarely Japanese drama. If I am playing a Japanese game with lots of dialogue, I maybe have like few hours of immersion every day, if not maybe between 30 and 60 minutes at most. Of course I know Anime or games aren’t great for learning how to speak Japanese, let’s put that aside for a now because I am mostly asking about vocabulary.

Now here is my question. I learn around new 8 words per day but I have been thinking, maybe I can lower that amount and instead I would focus on more immersion? Do you guys think this is a good idea? Can you even learn new words just by watching things? Or do I have to do some stuff like taking notes of words I don’t know then check their meanings. Or maybe since we use subtitles (hopefully) we learn it naturally? Or maybe our brain is so great that it can guess meanings of the words by context. Basically, do you guys improve your vocabulary just by watching Japanese things? Or maybe immersion is for improving long term memory about the words we have already learnt.

What do you think? Just talk to me about your experience please.

6 Likes
#2

From my experience with other languages, I’d say watching videos is a good idea, if they have subtitles on the language you are learning, then better, and if they are tailored to your level, or a bit more advanced, then it is also great. Watching a video where you don’t understand anything is not going to magically click at some point.

5 Likes
#3

Similar to the other response, the key to immersion is that you actually are understanding a solid chunk of what is being said - perhaps you need to really focus on it, which is fine, but if it’s sorta just becoming background noise, then I personally don’t think it’s doing much of anything. So immersion is great as long as it’s the right level, and I think it’s always worth making sure you’ve found material that really suits you.

Honestly I can’t say whether you should focus more on vocab or other learning sources - it’s a question I often ask myself. I sorta decided I would go up to the Bunpro N2 deck before maybe slowing down, so that’s 6k words. With other miscellaneous vocab words I have, my total would likely be around 6.5k by the time I finish. Just to give you an idea.

3 Likes
#4

Since you don’t mention it in your post I want to ask: do you like music? Because getting into Japanese music has definitely been beneficial for me expanding my vocabulary. Memorizing lyrics to songs you like and listen to often is, for me, a lot easier and effective than SRS.
Currently I listen to Porno Graffitti and RIP SLYME a lot, and would recommend those if you like rock / hip hop.

And if you’re open towards idol culture I would highly recommend checking out AKB48. They have a huge song catalogue, if you include all the related groups under the “48G” umbrella I think it’s over 1,500 songs.
AKB also has a massive amount of tv shows (variety shows, dramas, and documentary-style shows that dive a little deeper into the cultural aspects of different regions of Japan).
I’m currently watching through the full runs of AKBingo and AKB48 SHOW! (variety shows) and I’m finding it both very enjoyable and beneficial towards improving my Japanese.

I don’t mine words I learn with immersion though, not through any media (I already do vocab with Bunpro decks). I primarily look up words I don’t know in a JP-EN dictionary, and if it’s a word I find interesting or warrants more nuance I write it down and later look it up in 大辞林.
Then I just move on, being fine with forgetting words because if they are important enough to remember I’ll see them again later while consuming Japanese content, and the words will then get reinforced in my memory through continued natural exposure.
I feel this approach works great for me.

1 Like
#5

My thought is that both passive listening and vocab anki train the same muscle. Imo what’s missing in the repertoire are tasks that stress language production. My suggestion is to spend at least half your time doing:

  • Focused shadowing of your favorite media
  • Copying passages from memory
  • Journaling in Japanese (get a genkou sheet and write 250 chars)
  • Live conversation, if possible

Writing is especially useful since the characters help make sense of the approx 20 trillion Japanese homophones, but honestly any production task is beneficial.

I’ve found this to be the best way for me to acquire and retain vocabulary long term.

2 Likes
#6

I really… don’t like Japanese music. Especially Japanese hip hop sounds pretty bad for me for some reason… I watched some shows on Netflix but I can’t watch them without subtitles, not yet.

Thanks everyone for their input!

#7

"For dedicated study do the thing where you can keep you concentration closer to 100% + it’s harder + it’s interesting + more relevant to you + you can continue doing it for more time.

Try to become one with language and truly master it, writing and speaking as well"

At least it’s what I’m saying to myself now

1 Like
#8

I’m kind of biased with this sort of question, because I’m an “oldschool” (actually, just old :man_white_haired:) learner, and Anki literally didn’t exist when I first started out with Japanese.

So pretty much all vocabulary that I learned (outside of structured environments like classes) was from reading/listening to native content, looking things up in dictionaries myself, and occasionally writing stuff down in a notebook when I wanted to review it.

Nowadays, it seems the general consensus in the learning community is that Anki is a must for learning vocab efficiently, and I can certainly see the benefits, but I do know at least a few ‘modern’ learners who are doing it the way I did, and prefer just engaging with native content to learning vocab in a vacuum. As you might imagine, different people have different interests and different ways of learning.

So my advice would probably just be to go with what feels right to you. If you feel like you’re benefitting from your Anki use, then keep doing it. If you feel like it’s becoming more of a chore than it’s worth, don’t be afraid to put it down (and you can always come back to it if you find yourself missing it).

The only “right” way is the one that works for you because you find it meaningful and engaging enough to keep going with it. It may sound a bit cheesy, but in my experience it’s very much true.

4 Likes
#9

5,000 words!
Why are you studing Japanese? You have enough to do what’s ever you want.

You got it backwards - you don’t watch Anime to learn vocab, you learn vocab to watch Anime.

I think Sometime between 1,000 and 2,000 is when to try immersion. I’m at 5892 sentence in Anki - 3873 with Japanese definitions from thesaurus.weblio

I’ve read 2 dozen volumes of manga, a dozen seasons of anime, 5 movies and 30 pages of a kids novel [for 10 year olds]
I can turn on the news and read NHK [not NHK easy]

I’ve adveraged 5 cards a day in anki over the past 3 years.

OK, I looked it up https://www.researchgate.net/figure/Distribution-of-person-parameters-in-the-empirical-sample_fig3_301693773

In a survey of 471 kids in Germany, 5,000 is a 6 year old. a little low… 6,000 for 7 and 11,000 for 8 years old.
1000003922

1 Like
#10

You’ll want to look things up. Immersion alone is great for reinforcing what you already know, but it’s not as great for learning new things.

If you need to look something up multiple times, it might be a good idea to set some time aside to study it seperately (like adding it to anki).

In my experience, unknown high frequency words are good to study regardless because you’ll be running into them anyways. Might as well defog them preemtively.

1 Like
#11

Okay. Maybe you’d like to try to see if something like 乃木えいご is appropriate immersion for your level? It’s not fully subtitled, but it mostly is. The language use also isn’t very complicated.
It’s about girls from an idol group being taught English. It should be pretty easily digestible because the girls aren’t fluent in English, so they keep it sort of basic. It might also be interesting for you to get a change of perspective, look at language learning from a Japanese speaker learning English.

4 Likes
#12

Understatement of the year :sweat_smile:

2 Likes
#13

I read that you need around 9000 words in Japanese to understand most of the things. That makes sense because Japanese have so many words for same stuff, that is one of the hard things about it.

1 Like
#14

That looks like an interesting show! Japanese do really make some interesting TV shows.

1 Like
#15

But for reading books its just the start.
With the vocab of that that size you’ll still have to look up one word per sentence or so. But for human conversations and YouTube it sounds quite reasonable

#16

Ah, I don’t read at all. I don’t focus on Kanji, my goal is being able to talk and listen.

#17

Just thought I’d add some info which may help.

This paper focuses on what order learners should learn vocab in, for the sake of efficiency, but also it touches on the percentage of words you’d need to know to understand different domains (based off of corpus data) and also the clash between vocab order and kanji order that many learners face. The two things I want to highlight from it are the amount of vocab needed for coverage of the written language, and also how learning strategies for vocab may need to change at the intermediate and advanced level.

Have a look at the table on page 155 (and the surrounding chapter) for full details but just roughly taking averages, for coverage of written texts you need:

about 1,500 words for 80%
about 5,000 words for 90%,
about 10,000 words for 95%,
about 20,000 words for 98%
about 30,000 to 35,000 words for 99%

This is excluding things like names and function words (I think - been a while since I read the paper). Note that this is coverage for magazines/books/news. Coverage for internet forums is significantly lower; you can find the exact data in the paper itself.

Regarding how learning strategies may need to change as you progress: basically, when you are a beginner you can capture the highest frequency words from essentially any type of text, as they are unavoidably common. Once you hit an intermediate level, domain specific vocabulary becomes far more important for increasing comprehension. People normally feel this effect quite heavily when switching from a domain or type of media they have a lot of experience with to one where they have near zero, and this can lead to the ‘intermediate plateau’ feeling. Don’t forget that the numbers above are based on learning the most frequent words in the exact order of frequency - this is quite unlikely to happen for most people!

Next, NTT (a large telephone company) researches native vocab counts. Keep in mind that this is a measure of passive vocab when reading, not active spoken vocab. Below is their current estimation split by age:

中学生：20,000語～57,000語

高校生：32,000語～61,000語

18歳-20代：43,000語～74,000語

30代：58,000語～84,000語

40代：66,000語～91,000語

50代：72,000語～96,000語

60代：74,000語～100,000語

70代以上：76,000語～104,000語

You can read about the methodology here (in Japanese) or you can try and take the vocab test yourself here, although it is probably of limited use to foreign learners. Make sure you know the reading and meaning without simply guessing based on the kanji; you should have seen the word before in at least one context.

Regarding immersion and vocab, Paul Nation has a lot of research on vocab acquisition and the long and the short of it is that extensive reading is extremely powerful for learning vocab. The research on extensive listening is a lot more sparse. Both are important, though.

Note that the way words are counted not only differs between languages but also between research papers within the same language, so best not to get too focused on specific numbers.

5 Likes
#18

Just thanks for posting some actual research :slight_smile: I always love when people do that :slight_smile: