Hello everyone!
I’ve just watched this Youtube video about improving speaking skills with input training only. In the video, he said that he studied English for a year without training his speaking skills at all. When he tried speaking English with a stranger for the first time after a year of listening only, he was surprised by how fluent he was. What do you think? Do you agree? I’m trying to improve my speaking skills, but I don’t have the opportunity to speak with Japanese people, so I’m wondering if it can really work.
1 Like
As someone who has been learning for almost two years and has had almost zero output during that time, I can tell you that this isn’t my case. Lately, I’ve been using HelloTalk to talk with native Japanese speakers, and oh boy, do I struggle. I mean, I can understand what they’re saying to me, but whenever I try to reply in Japanese, I can barely form complete sentences. Sure input has help me a lot to understand different kinds of media but speaking is a different skill so Im currently implementing output on my daily studying.
You could try hellotalk or other similar apps to find Japanese speakers. There are many Japanese people that are looking for english speakers to practice as well so you can make a language exchange.
2 Likes
Well, I speak once every few months, and every time I’m like “wow I speak so much better!”. I feel I don’t really have to do much output to reach a high level, even as it’s now, my output abbilities are quite good, but my understanding is not even at a level of 18 y.o native yet, so I’m just working towards that point, the point where I can read books easily, and understand everyting in great detail. So I can feel progress in my output each time I take in a good portion of a language, starting from the begining.
On the other side, I’m helping (teaching) my granny learn Italin, and while her comprehension is at a certain level, she struggles soo much with output (even though we do a lot of speaking).
Again, my mom understands italian better then me, but I can see a lot of her mistakes that I already don’t make. But she’s not neccessery much worse at speaking than I am, and also her studying was based more on textbooks and courses, while mine on youtube and a few books.
All above, and some more thinking of mine, and I decided to think that age plays a big role in output, and less in input.
A lot of people who talk about input only, especially in japanese learning community, have learned Japanese while being pretty young, so a lot of them might no realise it doesn’t neccessery work like that for all groups of people.
The person in the video is also young, and getting good at output with only input was probably true for him.
Again, I think it’s possible to kind of develop some kind of passive language. I knew Ukrainian for a long time, from televition, school books, films, some other stuff, but have never used it, and when I would start to use it, it would’t sound to fluent even though I was native (I was thinking a lot about words choise, and now after 3 years I still do, now mostly for sounding good). Now my russian is like that, I haven’t spoken it for 3 years and when I start for some reason, I sound incredibly strage, and cannot find words for simple sentences without effort.
But to know if you do progress or not, you have to speak form time to time to catch that
2 Likes
Speaking in a foreign language is often an exercise in finding what you truly understand. The words and phrases that just come out because you know they’re meant to be there. To that end, lots of input is extremely helpful. Native input will help build your instinct for what sounds ‘right’ in a foreign language better than any other method.
However, the actual skill of speaking - turning off all the other processes in your brain and just going for it, is one that some people start off a lot more adept at than others. Nevertheless, it can be trained. I am not very ‘talented’ at this base speaking, and so it turns out that people who objectively should know less than me can more fluidly speak because it does not make them nervous the way it makes me nervous. The thing I need to practice in this case is speaking, not Japanese per se.
That’s kind of rambly and I think not very well communicated, but the base idea I’m getting at is there are two sort of distinct skills to consider in regards to the topic of ‘speaking in a foreign language’
3 Likes
Worked for me but I did practice, just not much at all. I had many thousands of hours of input before any real output with someone like 2 years later. Practice writing a few sentences here and there. Daydream in Japanese in the shower. You’ll do it regularly and not much else to do in your head while you’re there. This brought me to a level where I was able to communicate, just with a lot of anxiety and pretty clumsily. Making a friend and talking to her a lot brought me to a decent enough level. Not amazing, but that’s fine and kind of fun. I understand a lot, my output vocabulary is lagging behind a bit, but not so much I can’t explain myself around ど忘れing a word and move on instead of freezing.
I think a large amount of input and a little output here and there is fine unless your goal is to talk to people then just have fun talking to people once you can understand enough to respond with broken Japanese and take it from there, listening to your conversation partner is input too. 好きこそものの上手なれ
I think hours matters more than age. Based on my hours now I’ve probably progressed at nearly the same speed as when I learned English as a second language as a kid. I’m “young”, but like nearly thirty young not teenager young. Kids spend a lot of time goofing around in a language. Adults usually don’t. Of course a kid playing MMOs and talking to people 6 hours a day is going to improve fast, the kid is getting like 2000+ hours of varied input/output per year. Most adults might spend 1-3 hours at most because they have more responsibilities, less energy and want to relax. But that’s not the age talking yet. And I’ve met some 50 year old refugees who are amazing at picking up a new language too, they put in the hours.
2 Likes
OK, thanks. I’m actually already using HelloTalk, but I always feel more confident when it’s in real life. I think that more input for less output might work for some people, but not everyone.
I think it’s more about what you can produce vs what you know.
If you know a lot, but haven’t spoken in a while, when you start speaking you’ll be doing a lot of effort to remember everything you’ve already heard and make make it automatic, relearen to know make automatic mistakes you can see already.
But when you reach that point where you can say all you know with a good confidence, you start understanding that you don’t have enough of that yet passive knowledge, so you should go back to input to get more of that, so you now have something to internalise.
I think if you pay attention you can feel “oh okey, I speak a lot but don’t make any progress, I probably should do more input now”, and backwards. At least in my experience I can feel that, and based on that intuitive feeling I make changes in my learning.
Hm… I do feel it went that way for me in English, but I can’t say if it’s the same for Japanese.
I live in The Netherlands and my native language is Dutch. When I was a kid, I played video games in English, started reading a lot of things on the internet in English and at some point also started interacting with others online in (written) English. I did not have any experience speaking English until secondary school where I got English classes for the first time, and indeed, I was almost fluent from the get-go.
Some of the sounds that appear in English that are not native to me remained a little bit difficult to produce. This mostly disappeared when I got a British girlfriend later, travelled to the UK for several weeks at a time and was completely immersed in the language and had to speak it every day, all day. (Though some sequences of sounds are still difficult for me to produce even 20 years later.)
It’s definitely not that way for me in Japanese right now, but then again the amount of Japanese content I’ve consumed is much smaller than the amount of English-language content I consumed in the first 12 years of my life, and Dutch and English are closely related languages and I’m sure that accounts for something as well.
But all in all, yes I do believe it’s possible to gain fluency from just consuming input. I’m sure it’s possible to sufficiently piece the entire language together in your brain through sheer consumption that the only thing left to do is to then produce it orally. At that point, it might still require some ironing-out of sound production difficulties.
4 Likes
I ranted about this on my study log
We need to keep in mind time is a finite resource.
Starting point was me and two friends, all 3 with similar lifestyles and motivation levels. So we had roughly the same amount of time, let’s say 2 hours a day to simplify the argument, though of course it wasn’t two hours a day.
Both my friends went the classic optimized route: Anki recall (input-only), immersion
I went Bunpro cloze (production), preply tutor (output), immersion
But we all spent roughly the same amount of time “learning Japanese”, just the distribution was different.
So now whenever we’re in Japan and actually get into conversations, roughly 1.5 years after starting to learn:
My friends had substantially more input than I had.
I had substantially more output than they had.
Result: their passive vocab and comprehension is larger than mine, my ability to produce sentences and my spoken grammar control is larger than theirs. Who would’ve thought 
I don’t think there’s a better or a worse way here. Optimizations are mostly about what skill you get first, and which one you get later. And you can always reprioritize mid way: We met in Japan in June, I noticed their comprehension is much better than mine, so I upped my input by a lot and stopped using cloze for a while to save time. They noticed I can make spontaneous multi-clause sentences and express more nuanced thoughts while they fumble, now they got a preply tutor as well. Next benchmark will be November.
And besides, there’s no super strong need to optimize. Almost everyone here who is not a native English speaker, has learned English the most inefficient way possible: classroom lessons with 30 other children over the course of a few years with a demotivated school teacher. And yet, I get surprised when people like @BreadmanNin say they’re not native speakers. I doubt anyone in their school years sat down to figure out the best moment to start outputting English yet here we are using the language anyway.
6 Likes
I personally didn’t improve until I started speaking.
3 Likes
input does improve your speaking in terms of colocation knowledge etc but only more output will improve speaking automaticity… I read a lot and I still struggle a lot with automaticity. A lot of my sentences get cut halfway through or I use so many fillers because I am still thinking it through too much. I might choose more natural word choices (sometimes) but I think input improves output in specific (not insignificant) ways and leaves lots of gaps that can only be improved by output itself.
3 Likes
also another thing is that I have a much larger database of Japanese now in my brain but I have a much lower retrieval rate and knowledge got a lot more fuzzy so that has made my output ability even slower I feel lol
3 Likes
I think this is certainly true to an extent. Although, the main reason in my opinion, is that only input will actually get you to truly improving comprehension to a large degree (which is necessary for fluent conversations).
Practicing output is still necessarily to actually improve your fluid speaking ability, but it can be put off for a while until your foundation of the language is at a higher level. If you start practicing output once your base is already higher, it will go a bit smoother (in my experience).
Anyway, as others have said in this thread, it pretty much comes down to your priorities and what you want to do with your study.
3 Likes
I think all real progress revolves around output (I’m guilty of avoiding it the most), with the catch that the output doesn’t have to be said out loud, it can be a prediction/thought.
At master level I would want to be able to predict several branches of what you will say. This should take so little effort that I could already start choosing the best possible replies to these branches before you finish your sentence.
I would care about the funny replies the most, but these would rely on knowing all the common replies anyway.
I think this can only come from output practice and input can only provide some polish.
I agree with this mostly, but in my opinion predictive thought comes from lots of input and recognizing speech patterns and typical responses. I feel like without constantly consuming native content, it would be difficult to build an internal foundation on what people typically say in actual situations. Therefore, for this aspect specifically, I feel like it would require lots of output, as I’m not sure what type of practice one could really do to improve ‘predictive ability’ without actually consuming native content to build the intuition in the first place.
1 Like
Here’s the skill improvement matrix:
Practice speaking (with partner):
- Speaking up 10
- Listening up 8
- Reading up 3
- Writing up 4
Practice speaking (without partner):
- Speaking up 7
- Listening up 5
- Reading up 1
- Writing up 3
Practice Listening (pure):
- Speaking up 6
- Listening up 10
- Reading up 4
- Writing up 2
Practice Reading:
- Speaking up 5
- Listening up 5
- Reading up 10
- Writing up 6
Practice Writing:
- Speaking up 6
- Listening up 4
- Reading up 7
- Writing up 10
These are the definitive weights I decided, so conduct yourselves accordingly.
The difficult thing about this type of topic is all the science shows drastically different results based on many different factors. Your age, amount of active listening/reading, source and style of conversation.
Input is essential as a matter of fact. The amount of output however tends to be debatable among many different studies.
I personally think output is essential, and this can be spoken or written. However, only output with no form correction doesnt help. I think writing is horribly under represented when it comes to language learning, more so in Japanese too since “no one writes anymore”. And this is true, but writing forces you to read, forces you to slow down and think about form.
Input only means you may understand but not understand how to respond. It can however help with comprehension and sometimes you hear/read a piece of grammar and something clicks and now you better understand how to use it.
One thing to always try to remember though that a language is broken in 4 pieces, These four skills will never be of equal value even in the mother tongue.
Speaking
Writing
Listening
Reading
What I think often happens, especially in older learners (25以上) is that the small bits of output actually dont stop. We think and consider how to say certain things. So even though one may not be “outputting” per se, the seeds of thinking are still there. Listening and understanding and then making a new sentence based off what you just heard. So then when it comes time to output it seems amazing. All youve been doing is listening, and now you can speak! Amazing! When I dont think that is quite the case at all.