Input vs Output

Index » General discussion

 
Mcjon01 Member
From: 大阪 Registered: 2007-04-09 Posts: 551

I don't think the input over output argument says that you shouldn't ever have any output at all, or that you should just magically go from no Japanese to perfect Japanese.  I do, however, remember reading somewhere that you should only say things that you know how to say.

Tobberoth Member
From: Sweden Registered: 2008-08-25 Posts: 3364

Nukemarine wrote:

Samsara,

Technically my reviewing cards in Anki are both input and output.

For one type of card, I "hear" the statement in Japanese (input) and must "write" out the statement in Japanese (output). If I get the output wrong, I get immediate correction, as the answer shows the proper writing. In addition, there are definitions of terms in the statement, and an english translation if needed (usually not).

For the other type of card, I "read" the statement in Japanese (input) and must "say" the statement in Japanese (output). Again, if I get it wrong, I get immediate correction by the playing of the statement.  Once again, definitions and translations are also provided if needed (though normally not on mature cards).

Granted, I now view all the reviewing of cards, and the studying that went into creating the card differently. I look at it as making my real input more efficient. By real input I think of Manga, Music, Dramas, Conversations, etc. I hate forcing the output because I know it's horrible, and few people will bother correcting me.

That isn't really output though, you're not creating any Japanese, you're just mimicking it. I'm sure it helps both hearing and pronunciation, but it won't help you with being creative with Japanese since you're not actually producing anything. I could have several audioclips of french sentences and mimic those, it wouldn't mean I can actually speak any french, I would just be copying sounds.

There really isn't a good way to train production with flashcards. You need to write essays, speak to natives etc for it to really get going. Force yourself to use the grammar instead of just hearing and understanding it.

Dragg Member
From: Sacramento, California Registered: 2007-09-21 Posts: 369

McJon, many people here who feel strongly about extreme input-over-output are following the ideas of Khatzu, who, in turn, got them from from some other people.  As far as I can tell, Khatzu thinks that you should not output too early, even if you know the sentences.  In one of his AJATT articles he writes, "When you learn sentences in Japanese, do not force yourself to use them."

He doesnt advocate actually using any sentences until you do reach a sort of intuitive perfection because then you won't make any mistakes which could become harmfully ingrained.  He claims, "When the time comes for you to use those sentences, you'll know."  This is just my interpretation of his words, of course, but I think some people get a bit confused about exactly what he means because he does sometimes describe a transition point of being able to suddenly output perfectly after a long period of input with little or almost no output beforehand.

Last edited by Dragg (2008 September 08, 1:02 pm)

Advertising (register and sign in to hide this)
JapanesePod101 Sponsor
 
leosmith Member
Registered: 2005-11-18 Posts: 352

Zarxrax wrote:

According to Khatzumoto and various other sources, there is no need to work on your output.

Conversation is, arguably, the most desired skill in a foreign language. It's also, arguably, the most difficult. I like to start torturing a language partner after knowing a few hundred words and grammar patterns. Speak early, speak often. That's my motto.

No matter how long you wait, it's going to be difficult, and you will make many, many mistakes. Those who practice frequently, and who don't worry too much about messing up will learn the fastest and the best, in my experience. I used to be really sensitive about making mistakes, and it hindered me. The more languages I learn, the less sensitive I am, and the more I realize hiding behind a book/computer all the time is a mistake.

How about that Khatzumoto? Didn't he brag about how much he talked with native speakers when he was learning? This is just one more thing he did that he recommends others don't do.

mystes Member
From: USA Registered: 2008-04-08 Posts: 99

Tobberoth wrote:

That isn't really output though, you're not creating any Japanese, you're just mimicking it. I'm sure it helps both hearing and pronunciation, but it won't help you with being creative with Japanese since you're not actually producing anything. I could have several audioclips of french sentences and mimic those, it wouldn't mean I can actually speak any french, I would just be copying sounds.

There really isn't a good way to train production with flashcards. You need to write essays, speak to natives etc for it to really get going. Force yourself to use the grammar instead of just hearing and understanding it.

I think it is important not to conflate production and output. For being able to employ grammar properly (at least in the sense of using it grammatically correctly) practicing production by using flashcards is actually sufficient, in my opinion. On the other hand, I consider activities such as writing essays to fall under the category of (true) output. I don't believe it is necessary to practice output to practice production of grammar. The advantage of this is that it is that it is possible to postpone the more complicated issues of production until after developing an ear what is correct and what is not. For example, I think writing essays from day 1 is rather counterproductive. Instead, it is better to wait until you can read essays to learn how to write them correctly. Consider this description of how to read carefully when learning English from antimoon.com:

1. Stop at interesting (not obvious) things: a new word, how a word was used, a grammatical structure, a preposition, an article, a conjunction, the order of words, etc. For example, spend a while to think about the fact that the sentence contains the preposition at, and not on. Perhaps the sentence uses the present perfect tense where you would have expected the past simple. Perhaps the word order is different than in your first language.
   2. If the sentence contains a useful phrase, ask yourself: Could you produce a similar phrase yourself? Would you use the right tenses, articles and prepositions? Would you use the right word order?

I think this is good advice for learning Japanese, as well. The problem is that this sort of careful reading is impossible until you have attained a sufficient level. That is why you have to do a lot of memorizing first.

I don't know if this is optimal, but I don't see there being many other options without receiving corrections from a teacher. Just churning out essays riddled with mistakes isn't going to improve output, in my opinion. Of course, with the increased access to native speakers the internet has provided, it may not actually be that hard to find someone to provide corrections.

Last edited by mystes (2008 September 08, 2:44 pm)

Tobberoth Member
From: Sweden Registered: 2008-08-25 Posts: 3364

Just as a quick answer to the previous post, yes, i did mean write essays which are in turn actually corrected. All form of output should be verified by a teacher, a native or at least a fluent speaker of the language. All forms of output which is not corrected is just an attempt at Japanese. It's output, but there's no telling if it's proper or not, which is why I personally don't recommend it.

Samsara Member
From: USA Registered: 2008-06-30 Posts: 33

Mcjon01 wrote:

I don't think the input over output argument says that you shouldn't ever have any output at all, or that you should just magically go from no Japanese to perfect Japanese.  I do, however, remember reading somewhere that you should only say things that you know how to say.

I don't remember if it was on this forum or another one, but recently I read a post where someone asked for advice on how to use the "no output" theory in her Japanese class. Her professor wanted her to practice pronunciation and speaking in class, but she didn't want to practice output "too early."  So yeah, there are some people who take "input before output" to mean that they're not going to speak a word of Japanese for years... and that is just plain crazy.

@thermal: Thanks for the suggestions on bridging the gap between mimicking and actually producing.

iSoron Member
From: Canada Registered: 2008-03-24 Posts: 490

QuackingShoe wrote:

The solution to your brain being lazy is to stop letting it be lazy. When an adverb uses と, (for example, 意外と), pay attention. When people use を with 言う (or 思う or such) instead of と because they're not actually quoting but rather talking about what they're saying, (IE 冗談を言う), note it.

Let us consider two kinds of cards.
Which one do you think is more effective?

1. Recognition card

    Question side:
        - 私は窓を開けておく。

    Answer side:
        - わたし は まど を あけて おく。
        - I will open the window. (for future convenience)

    Thinking process:
        - Pronounce the words. Ok.
        - Now let's see. 「私は」 Ok, nothing new.
        - 「を」 Ok, direct object. Makes sense.
        - 「開けて」 Te-form. Nice.
        - 「おく」 Oh yeah, we use this in this situation.
        - Would I be able to build this sentence? Yeah, probably.
        - Next card.

2. Production card

    Question side:
        - 私、 窓、 開ける
        - I will open the window. (for future convenience)

    Answer side:
        - 私は窓を開けておく。
        - わたし は まど を あけて おく。

    Thinking process:
        - 私は...
        - 窓... を? Yeah, it's を.
        - Now let's see. What's that rule for 'future convenience' again?
        - Oh yeah, 「~ておく」.
        - 開ける... 開けて.
        - 私は窓を開けておく。
        - Yey, correct. Next card.

General output is too broad to be used with an SRS, but this kind of limited output can be used just fine. Of course, it is only applicable to certain situations (grammar rules, particles, exceptions, etc), so it doesn't replace the sentences method + pause and think. But it's a nice complement.

phauna Member
From: Tokyo Registered: 2007-12-25 Posts: 500 Website

I'll throw in a couple of ways to output.

Question:
週末、なにした?

Answer:
You answer the question aloud.

ーーーーーーーーーーーー

Question:
青が一番好きだよ。  (??)

Answer:
You ask the question aloud.

Simple ways but they get you thinking.  Also, it mimics a conversation somewhat.  Any question cards can easily be converted into the first kind of output card, ie you answer the question.

thermal Member
From: Melbourne, Australia Registered: 2007-11-30 Posts: 399

Tobberoth wrote:

I don't want to mark words, but I guess the point against output which Kazumoto makes is that you might start using unnatural japanese. Just to have some example, "なんか健康的がいい" is probably not something a japanese person would say. Think about it. "なんかおいしいがいい". You're commenting on the actual adjective, not on the thing you're talking about. A japanese person would probably say "健康にいい物" or "健康的な物".

However, I do sort of disagree with Katzumoto since I do not believe you can go from speaking no Japanese to speaking perfect Japanese. We aren't natives, it's okay to speak unnatural Japanese while learning, as long as you're careful and learn from your mistakes. Just thought I'd bring that aspect of the argument up.

You are probably right about this sentence and I agree you have to learn from your mistakes. When I speak to myself I do pay attention to what I say. Sometimes I will say the same thing 3 times, until I find a way to say it that doesn't sound strange.

However, I think the issue of not having it corrected is something largely inherent with real communication. When I speak to my friends in Japanese they almost never correct me and I don't feel comfortable trying to force them to. Thus speaking to yourself is still something like 70% as good as a real conversation.

I also think that even making mistakes that you don't notice is useful. I think humans fundamentally learn from mistakes. However, it is still effective if you make the same mistake 50 times before learning the better way. This way you already have an existing framework (the way you are using that particular point of language) to amend rather than trying to magically imprint some principle in your mind.

Last edited by thermal (2008 September 09, 5:44 am)

thegeelonghellswan Member
Registered: 2008-05-15 Posts: 74

I think a lot of people are misinterpreting Khatzu's view on input and output. There definitely must be output, but when? From day one? like university classes where you blunder through and are judged on pronunciation when you have barely even heard the language? What is more natural is to begin outputting; that is, speaking and writing after long and deep exposure. For some this may come quicker than others. Yes there still will be mistakes, it is unavoidable and it is foolish to think that you can be a fluent speaker without having spoken before, but I firmly believe in input before output. I think as a rough guide if you are having to guess how to make/say a sentence you probably need more input, if you are intuitively producing a sentence based off hundreds/thousands of other sentences you have experienced, then you are probably on the right track.

I am paraphrasing, but I believe he said something like if you are wondering if you are fluent or not yet, you are not. So if you are wondering if you need more input or not, you do.

This is my take.

mentat_kgs Member
From: Brasil Registered: 2008-04-18 Posts: 1671 Website

I believe you should not avoid output. When you are in a situation where it is fun to speak japanese, with friends, for instance, you should speak japanese. But you should not force it in an alien way. RTK and the sentences are alien enough.