nest0r Wrote:We need to hammer down the definition for fluency so we can know how to get fluent in one year! We know that there are fluent people out there, they say so in their blogs and tell us they did it very quickly, but we need to *know* what they mean by 'fluent' so we can know if we're learning quickly enough to brag on the forum. When will I be fluent? Is this person fluent? What is fluent? This important.

Or you can just make a thread that says:

"I'm fluent after one year.

Discuss."

People too ignorant to know better or too stubborn to give up the garbo combined with ceaseless arguments started in your name will skyrocket you to fame in no time .

Using this method, after studying Japanese for one year, I am completely fluent*.

*In English. Individual results may vary based on English fluency at beginning of Japanese study.

nest0r Wrote:We need to hammer down the definition for fluency so we can know how to get fluent in one year! We know that there are fluent people out there, they say so in their blogs and tell us they did it very quickly, but we need to *know* what they mean by 'fluent' so we can know if we're learning quickly enough to brag on the forum. When will I be fluent? Is this person fluent? What is fluent? This is important.

haha never believe anyone on the internet, it's like the first rule of the internet.

I have met some school high school kids from Thailand and Singapore who became fluent (to my definition) in about 3 years after moving to Japan in their teens, but that's actual immersion (as opposed sitting watching dorama) and their kanji was pretty terrible apparently. I'm guessing they had special Japanese lessons as well.

Another hapless forum-goer falls victim to nest0r's infamous sarcasm attack.

日本語お上手ですね

According to my Japanese textbook, the appropriate response is:
うそです。 (that is a lie)
No, I lied. The real response is
そんなことではありません。 (no, that isn't true) or
いいえ、まだまだです。 (no, I have a long way to go)

So if a Japanese person says something like this to you, say either of these in reply and see what they say.

Back to seriousness. How do you measure fluency? The best method I have heard of was from Krashen who is a language researcher. He suggests interviewing and recording the test subject without their knowledge. The reason for this is twofold: The subject might speak differently if they know they are being tested, and secondly part of the fluency definition is speaking "easily and naturally" which can't occur if the person is speaking with the knowledge their performance will be graded. Once you have your recording, play it back for several native speakers. The native speaker should try to judge the subject's country of origin, and you should have at least one male and one female judge if at all possible. From the results, we can get a good idea of the subject's fluency. This method of judging fluency is not at all practical. However, I believe it is very accurate.

jcdietz03 Wrote:How do you measure fluency? The best method I have heard of was from Krashen who is a language researcher. He suggests interviewing and recording the test subject without their knowledge. [...] Once you have your recording, play it back for several native speakers. The native speaker should try to judge the subject's country of origin.

I still prefer this modification of the Turing Test:

elzr.com Wrote:Here’s a (controversial) idea for a language test inspired by the famous Turing test for artificial intelligence: a native speaker of language X engages in conversation with two other parties, one a native speaker of language X and the other a student of language X as a foreign language; if the judge cannot reliably tell which is which, then (and only then) can the student be said to speak language X. [...] The problem is that ridiculously few people would pass it if it where applied today. And because it seems impossibly difficult most people turn away, dismiss the test as wrong or irrelevant, and sink their heads in the sand (“what shouldn’t be, can’t be right”). Which only highlights the current sorry state of language education.

http://elzr.com/posts/the-ttoefl-the-tur...n-language

I agree with the Turing Test idea, but this could bring up huge hurdles with accent problems. If you have a horrible accent, then yeah, you should probably work on it, but accents can be a really tough thing to nail perfectly. Even if the words you say are identical to a native.

I think the "can you pass for a native speaker" idea is stupid for several reasons -- first, it reduces fluency to a binary proposition where you either "pass" or you don't. Language ability doesn't work like that. The whole idea that you can point to people and say "fluent" or "not fluent" is patently absurd.

Second, the ability to pass for a native speaker is useless to most people. You can be very proficient in a language and not pass for a native speaker, and in any case, it only makes sense on the phone or Internet or some place like that when they can't see you.

Third, it puts the entire language proficiency down to conversational Japanese, and as Asriel indicated, it puts a lot of weight on accent. Accent and pronunciation certainly is important, but you don't need to have 100% native speaker-level accent and pronunciation to communicate smoothly.

Personally I don't like abstract definitions of fluency or proficiency. To me, it just has to do with how well you can do things that matter to you in the language. The less of a barrier the language presents to you, the more fluent or proficient you are.

I just don't see what the point of defining this is. If you're able to talk to a Japanese person or read a book that interests you, that's more important than whether you can call yourself "fluent" under some abstract definition.

yudantaiteki Wrote:I think the "can you pass for a native speaker" idea is stupid for several reasons -- first, it reduces fluency to a binary proposition where you either "pass" or you don't. Language ability doesn't work like that. The whole idea that you can point to people and say "fluent" or "not fluent" is patently absurd.

Second, the ability to pass for a native speaker is useless to most people. You can be very proficient in a language and not pass for a native speaker, and in any case, it only makes sense on the phone or Internet or some place like that when they can't see you.

Third, it puts the entire language proficiency down to conversational Japanese, and as Asriel indicated, it puts a lot of weight on accent. Accent and pronunciation certainly is important, but you don't need to have 100% native speaker-level accent and pronunciation to communicate smoothly.

Personally I don't like abstract definitions of fluency or proficiency. To me, it just has to do with how well you can do things that matter to you in the language. The less of a barrier the language presents to you, the more fluent or proficient you are.

I just don't see what the point of defining this is. If you're able to talk to a Japanese person or read a book that interests you, that's more important than whether you can call yourself "fluent" under some abstract definition.

I like this post

I agree 100%

yudantaiteki Wrote:I think the "can you pass for a native speaker" idea is stupid for several reasons -- first, it reduces fluency to a binary proposition where you either "pass" or you don't. Language ability doesn't work like that.

There's no reason why there couldn't be multiple versions of the test, testing different levels of proficiency. Some variations I can think of, in ascending order of difficulty:

JLPT-style: Have both native and non-native speakers take a multiple choice test. Non-natives that perform as well or better than the native speakers pass the test.
Composition: Have both native and non-native speakers write a text about a given subject, under given rules. Judges try to determine which texts were written by whom.
Text chat: Have both native and non-native speakers chat with the judges over IRC, a web forum, or some similar text-chat platform. Judges try to determine which chatters are the native speakers.

Quote:Second, the ability to pass for a native speaker is useless to most people.

You can't really say you talk the same language as everyone else if everyone else can easily tell your language apart, can you? You can say you speak something close enough to Japanese, and that's fine, but not Japanese.

Quote:Third, it puts the entire language proficiency down to conversational Japanese

There's no reason why the judges can't ask questions about other aspects of the Japanese language. They can ask the participants to talk about kanji etymology while speaking in Keigo; or test their vocabulary in a Shiritori match where only vegetables names are permitted. Of course, they'd have to keep it at a reasonable level, or else even the native speakers would have a hard time doing the test.

Quote:and as Asriel indicated, it puts a lot of weight on accent.

I wonder why we think this is. I guess it's only because accent is an aspect we language learners tend to be particularly bad at, and are usually forgiven. If we dreaded kanji, then people would complain about how the written test puts a lot weight on these little symbols. If we aced accent, then people would complain about how the test puts a lot of weight on the cultural aspects of the language (judges could ask you about an expression that was quite popular some ten years ago because of some TV commercial), etc.

iSoron Wrote:]
There's no reason why there couldn't be multiple versions of the test, testing different levels of proficiency. Some variations I can think of, in ascending order of difficulty:

That doesn't change my basic objection that proficiency in a language is not pass/fail.

Quote:You can't really say you talk the same language as everyone else if everyone else can easily tell your language apart, can you? You can say you speak something close enough to Japanese, and that's fine, but not Japanese.

So is that what you're going to say then? If someone asks "Do you speak Japanese?" you'll respond "No, I speak something close to Japanese, but not Japanese."

And in any case, that still doesn't really answer my question -- why is it useful to be able to pass for a native speaker in instances when the person can't see you?

Quote:

Quote:and as Asriel indicated, it puts a lot of weight on accent.

I wonder why we think this is.

It puts weight on accent to an absurd point. If you are able to speak Japanese that any Japanese person can understand, you don't need to have pronunciation so perfect that you are indistinguishable from a native speaker.

Quote:If we dreaded kanji, then people would complain about how the written test puts a lot weight on these little symbols.

The kanji analogue to this test would be that if you know all 1945 Jouyou kanji, you're fluent/proficient. If you don't know them all (whether you know 1944 or 2), you're not fluent/proficient.　 (Although even aside from this, if someone tried to claim that testing kanji, and nothing else, was a way to determine someone's Japanese fluency or proficiency, I would say that it puts too much weight on the "little symbols".)

Once again, I just don't see the point of such a test.

yudantaiteki Wrote:And in any case, that still doesn't really answer my question -- why is it useful to be able to pass for a native speaker in instances when the person can't see you?

Because if you can't, it means that there is something very distinctive about the language you use; which means that the language you use is not really the same language native speakers use; which means that you can't really speak the same language native speakers do. If that was your original goal, if speaking Japanese was your original goal (where Japanese is defined as the language the average person in Japan speaks), then it means that you have failed your goal.

If speaking Japanese was not your original goal, then whatever.

So according to you, even though I communicate orally with Japanese people every day and have very few problems, I cannot speak Japanese because I (probably) can't pass for a native speaker. My Japanese friends who are able to communicate in English with me, with few problems, cannot speak English because I know they wouldn't pass for native English speakers.

If that makes sense to you, I guess we don't really have anything further to discuss; your definition of fluency has nothing to do with actual ability in the language.

Being indistinguishable from a native speaker would probably put you at native level. Fluency is a different concept, defined by how fluent you are.

Eikyu Wrote:Being indistinguishable from a native speaker would probably put you at native level. Fluency is a different concept, defined by how fluent you are.

Wish my Japanese was at that level. Although my English is indistinguishable from a native speaker.

yudantaiteki Wrote:So according to you, even though I communicate orally with Japanese people every day and have very few problems, I cannot speak Japanese because I (probably) can't pass for a native speaker.

Yes. To communicate with the Japanese people you don't have to speak Japanese, you just have to speak something that is close enough to Japanese. People from Brazil can communicate rather well with people from Spanish-speaking countries. That's because Brazilian Portuguese is close enough to Spanish, not because Brazilians speak Spanish.

Calling Henry Kissinger and Governor Terminator, who are definitely not native speakers of English, "not fluent in English" seems to be a bit extreme. Though, the term "fluent" is very subjective and Zen, it's all inside you, so everyone has their own idea of fluency, I guess. I like this scale US government agencies use when I want to be more objective because it's pretty detailed and empirical.

http://www.govtilr.org/skills/index.htm

I usually call someone "fluent" when he/she is at the speaking level of around 3 in this scale, though the scale itself demands level 4.

Anyway, it sounds like what the article is calling "a seamless transfer of ideas" is more like "knowledge and sense of collocations of basic vocabulary" to me. And while it is definitely crucial, other skills needed to speak fluently, like the ability to instantly come up with appropriate vocabulary for a given situation, seem to be missed out from the article.

iSoron Wrote:You can't really say you talk the same language as everyone else if everyone else can easily tell your language apart, can you? You can say you speak something close enough to Japanese, and that's fine, but not Japanese.

I don't think it is fair to say a thing like this. Growing up, people often thought I had an accent and would ask me where I was from, despite that I grew up and was from where I was living. I don't know why I had an accent (I think maybe from reading so much).

So my language was clearly apart from everyone else's, but it would be completely untrue to say I am not a fluent English speaker (my only language then).

Native speakers of my native language, including where I grew up, often tell me that I have a foreign accent and use foreign-sounding language structures. How long till I'm fluent??

Of course by 'where I grew up' I mean the abstract digital realm where all bots grow up.

iSoron Wrote:

yudantaiteki Wrote:So according to you, even though I communicate orally with Japanese people every day and have very few problems, I cannot speak Japanese because I (probably) can't pass for a native speaker.

Yes. To communicate with the Japanese people you don't have to speak Japanese, you just have to speak something that is close enough to Japanese. People from Brazil can communicate rather well with people from Spanish-speaking countries. That's because Brazilian Portuguese is close enough to Spanish, not because Brazilians speak Spanish.

OK, have fun not learning Japanese, then. I guess this site should be renamed "Reviewing the symbols that are close to kanji" and then the forum sections should be "The language resembling Japanese" and "Learning something close to Japanese".

(I'm going to pretend that you don't honestly think the relationship between Brazilian Portuguese and Spanish is the same as the relationship between accented Japanese and native-speaker Japanese.)

People from England don't speak English.
Because seriously, who has "trousers" these days? Yes, I would like a pair of pants please, leave the "trousers" in your own country. We speak English here.
Come back when you learn to speak something other than "a language resembling English."

On the side note, people from Tokyo area can't even communicate very well with old folks from 東北 or 沖縄 who speak hardcore dialects, but they are still native Japanese speakers. And there is nest0r who speak C++ and Java with an accent...

mmm, my head will pop...

It's interesting, sincerely, that this debate is EXACTLY the same as the ones Japanese guys often have over what 英語ペラペラ means. Some say you have to be indistinguishable from native speakers, and some see people holding "sup?" "doing all right" kind of conversations and go like すごーい英語ペラペラなんですね！！！

Every time somebody asks me 英語ペラペラなんですか？ I have a headache and go like えーっと…　英語ペラペラかどうかはわからないけど、英語ペラ、位かな…　

It's about time I came up with a better line than that!

I think fluency, well you can call yourself fluent when you compare it to your native language. Can you write the same amount equally? Same with speaking,reading and listening?

You don't need to know everything in the language and be able to speak about it perfectly/write about,etc. I can't really talk about politics/economics because I'm not studying that or I just don't know about it well enough. But I can still call myself a native-speaker/fluent. It's strange but I think the best way is to compare yourself in your native-language. If you can do everything you can do but in Japanese your set. Get some native-speakers to check on your writing/reading/speaking,etc when your at the level to call yourself fluent to double check. But in the long run, the only one that will be able to know your fluent is yourself. (Sure there are some ways like tests but that can only go so far.)

I can speak Python with a severe Lisp.