Splatted Wrote:The sentences are:

博士はその道の権威（けんい）です。(This one is from memory, so may be completely wrong.)

彼女は臨時の店員（てんいん）です。

Thanks for any help.

Haha, one of my friends was dating a guy named けんいち, and could not pronounce his name the whole time they were dating. She started to just call him "Ken", and he made fun of her about the whole thing a lot. She, of course, could only pronounce it けんにち.

I can actually pronounce it quite well, but it's hard to explain how to do it. I came up with a lot of explanations in my head, but then realized they were all too off to be worth writing.

I think basically if you just remember that 'ん' isn't actually 'n' (standard romanisation) you'll be fine. English 'n' is made by lightly pressing the top of the tip of the tongue against the flesh just behind the upper teeth. For 'ん', the tip of the tongue shouldn't move from it's default position (as it would for な行), rather the sound comes from what the back of the tongue is doing. This will manifest itself in a few different pronunciations depending on the word, but I don't think it's necessary to think of it as 3 or 4 different sounds depending on the following consonant because this just complicates the issue.

nest0r Wrote:Learning pronunciation would come incidentally from developing a mental corpus and native-like intuition as as result of close/extensive (condensed) listening, speaking, and subvocalization, with occasional consultation for the ever-rarer instances of problematic interpretation of phonology.

Is this theory or have you actually done it? You can listen to all the audio you want, but if you happened to estimate a sound between L and ら or just stuck with the English approximation of い, it's unlikely you're going to move your tongue back or spread your lips more without the proper awareness. And let's say your ears were sharp enough to do it, how long would it take? 100 hours of listening? Would that be more efficient than spending a short amount of time reading about and practising the sounds before starting 100 hours of what would become reinforcement and further practice?

Maybe I've misunderstood your point (I'm assuming you've read something about it somewhere), but I would've loved a nice pronunciation book with a CD that covered everything from ラ行 and mouth/lip/tongue position for vowels to pitch accent and intonation. It's definitely possible to advance using your ears, but I did a ton of shadowing and was still mispronouncing ん, and I was never confident that my ラ行 was correct, even if it sounded good. Don't get me started on my trouble learning pitch accent. To have something telling me definitively what I should've be doing from the outset would've been invaluable. I can't say I sound like a native now, so I'd still be interested to read something that went into a lot of detail with practice exercises and what not.

nadiatims Wrote:I think basically if you just remember that 'ん' isn't actually 'n' (standard romanisation) you'll be fine. English 'n' is made by lightly pressing the top of the tip of the tongue against the flesh just behind the upper teeth. For 'ん', the tip of the tongue shouldn't move from it's default position (as it would for な行), rather the sound comes from what the back of the tongue is doing.

No. ん can be like the "n" in English depending on the position it appears in. What separates ん+vowel from なにぬねの is how you treat them in terms of mora.

For example, the reason 兄 (あに = ani) and 安易 (あんい = ani) are completely different to a native Japanese speaker' ear is that the former is of two moras while the latter has three moras. Replacing the nasal vowel ん in あんい with the English "n" doesn't make it sound like あに if the other aspects are all correct. It's still あんい, though it would sound odd or heavily accented. But ignoring how moras work might lead to something beyond odd or funny pronunciation; あんい can become heavily accented あに or incomprehensible gibberish which doesn't sound like Japanese.

Words like けんいち is hard because you need precise control over moras and good articulation of ん at the same time to get decent pronunciation. If you write the name from the viewpoint of how English syllables work, it's "kenich," which is the exact same syllable representation of けにち. But けんいち is analyzed as ke-n-i-chi in terms of mora, and けにち ke-ni-chi. Of course, the articulation of "n" is different between けんいち and けにち. But the wrong articulation of ん alone doesn't turn it into けんにち or けにち.

nadiatims does have a point in the latter half of his post though because correct allophone use and precise articulation are very minor if it's not important for you to pass for a native speaker born and raised in Japan. As long as you get moras and pitch patters right, the rest is quite minor and doesn't interfere communication too much. I don't think learning the 6 different variations of ん is the most important thing. It sure makes your Japanese pleasant to native speakers' ear and easier to understand. But most beginners have a bigger problem than that.

Anyway, I don't think it's necessary to learn about phonology and stuff extensively either. It's like grammar. Textbooks for beginners only give extremely simplified explanations of sounds, which is reasonable. And if you want precise descriptions, you have to learn a humongous number of rules encrypted by incomprehensible jargon, only to find as large a number of exceptions. On top of that, learning about phonology/phonetics alone doesn't improve your speaking or listening much; what you really need is a massive amount of exposure to native material (and practice for speaking). In fact, lots of people have achieved very high level without learning phonetics or phonology.

nadiatims Wrote:This will manifest itself in a few different pronunciations depending on the word, but I don't think it's necessary to think of it as 3 or 4 different sounds depending on the following consonant because this just complicates the issue.

This is how I see it as well. Learners are made aware that there are different sounds and listen to examples to ensure that they can actually hear the difference. Then it's just a matter of learning the sounds of words rather than attempting to apply pronunciation rules. The rules will be internalized without explicit instruction. In fairness to Magamo, I suspect he mentioned the rules as explanation rather than something to memorize and apply. [edit: wrote this before seeing M's post above]

As for the audio grammar stuff, it seems to me that the issues of using audio for grammar and audio for pronunciation have gotten a bit merged. They overlap, but aren't the same. Not really sure what bounaparte originally had in mind, but I don't think adding audio to text grammar references is the best way to go about improving either.

This seems to me like more non-issues which generate debate because of the way they are initially mischaracterized as a crisis or an either/or situation.

nadiatims Wrote:Wouldn't it be hugely beneficial at the beginner (still learning grammar) stage if all the decent grammar reference books were in digital format with all the sample sentences read by native speakers? this way you'd be learning correct pronunciation from the outset,

If someone's starting off in Japanese (or any language) by reading a dry, dull grammar reference from cover to cover, then that's their first mistake and it's no wonder they can't find any decent audio samples for what they're trying to learn.

nest0r Wrote:Indeed. I'm surprised that you had to ask whether I'm serious, or are rhetorically asking to express disagreement rather than stating something more straightforward. There's no need for extensive, separate pronunciation courses.

Extensive, no. But I think you can get a lot of benefit out of some very brief explanations and simple pronunciation exercises.

This is another area where I don't think the "audio exposure will sort everything out" works -- I've seen *professors* with decades of experience, including many years living in Japan, who still do not pronounce the ん + vowel sequence correctly. I didn't do it correctly after living 2 years in Japan.

For many people, once they establish their initial pronunciation, it's not going to change by itself.

Internet language fora are indeed very interesting places.
(fora is a plural form of forum)

Please forgive me my survival English.

I will try to explain why I need grammar right from the beginning.
When I start learning a new language, there are two things I concentrate on - and I find them very important:
1. listening comprehension
2. pronunciation

To do both 1. and 2. properly I need some kind of logic behind them:
I must know what kind of sounds there are, how they differ from the ones I already know, what phonetic features I must pay attention to while listening, etc.
If I don't know that pitch accent is important, or that Japanese spoken words are divided into morae (or moras, if you prefer), or that there are whispered vowels there, then I am bound to fail to notice them myself and substitute them by something completely different.

I must know what grammar features there are, how they differ from the ones I already know.
I must know that nouns have no gender or plural forms, that there are no articles, that the sounds -mas- carry the meaning of some kind of a polite form, that -u is the present/future tense, and that -ta is the past tense, that tenses are not only a characteristic of verbs but some adjectives as well.
I'd rather know straight away that おはようございます　is in fact　お早う御座います　and that they are forms of 早い　and 御座る, and that there's no 'good' or 'morning' in it.
I'd rather know straight away that どうぞよろしくお願いいたします are in fact forms of 　宜しい and 願う　and 致す and that 宜しい is a honorific form of 良い　いい　and 　致す　いたす　a humble form of する.
I'd rather know straight away that です　desu is a form of であるde aru and that である is made of で and the verb ある that is irregular and its negative form is ない/無いnai and that ない is an i-adjective and no longer a verb.
I must understand straight away that じゃない　ja nai is not something mysterious at all, that じゃ is in fact a phonetic contraction of で + は (わ) and that は here is in fact a topic marker and that the same phonetic contraction is to be found in 死んではいけない　sindeha ikenai, 死んじゃいけない　sinja ikenai. Etc, etc.　

Then I don't have to treat every single expression as something completely unrelated to other expressions/words/forms, and be puzzled all the time by something that it is not puzzling at all but only made so by bunglers or (not so) learned morons who write language handbooks.
And that saves me HELL of a lot of time later while dealing with authentic materials for native speakers right from the beginning. I prefer books I already love and know well.

It now should be obvious why I need a good reference grammar with good audio by native speakers.
By the way, grammars don't have to use artificial sentences - there are grammars that only use authentic natural sentences. Of course, if you don't like Miss Grammar, it is your business, not mine.

buonaparte Wrote:Internet...[snip]...mine.

That is exactly how I feel about it as well. It is of course entirely personal -I can understand that this does not work for everybody, to understate things-, but I also like to know all information there is. To the bottom. Complexity, completeness, they create extra links and make it easier for me to remember anything. And I find it very valuable in the long term. It may also help that I'm interested in language's structure itself; so thoroughly doing things, although slowing me down on the short term, I don't mind.
Oh, and I do not think that this is overcomplicating things at all, as you wrote below (I reckon in the spirit of the general opinion). Rather, the reverse: it prevents overcomplication later on. For me.

By the way, your links were very useful.

Furthermore, it isn't very helpful to discuss whether this approach is useful to some specific person, or has a point for him/her, or makes sense, as long as there are some who receive benefit of it. I'm glad that a topic like this receives replies varying from 'just approximate the myriad of sounds with this one sound...' to 'the phonology rules are like this...'. Pick what appeals to you!
Lastly, sorry, this spun a little bit off-topic.

Javizy Wrote:

nest0r Wrote:Learning pronunciation would come incidentally from developing a mental corpus and native-like intuition as as result of close/extensive (condensed) listening, speaking, and subvocalization, with occasional consultation for the ever-rarer instances of problematic interpretation of phonology.

Is this theory or have you actually done it? You can listen to all the audio you want, but if you happened to estimate a sound between L and ら or just stuck with the English approximation of い, it's unlikely you're going to move your tongue back or spread your lips more without the proper awareness. And let's say your ears were sharp enough to do it, how long would it take? 100 hours of listening? Would that be more efficient than spending a short amount of time reading about and practising the sounds before starting 100 hours of what would become reinforcement and further practice?

Maybe I've misunderstood your point (I'm assuming you've read something about it somewhere), but I would've loved a nice pronunciation book with a CD that covered everything from ラ行 and mouth/lip/tongue position for vowels to pitch accent and intonation. It's definitely possible to advance using your ears, but I did a ton of shadowing and was still mispronouncing ん, and I was never confident that my ラ行 was correct, even if it sounded good. Don't get me started on my trouble learning pitch accent. To have something telling me definitively what I should've be doing from the outset would've been invaluable. I can't say I sound like a native now, so I'd still be interested to read something that went into a lot of detail with practice exercises and what not.

I think many of us or most of us have done it, and pretty quickly and easily. I assumed that was the default based on previous conversations. Minor supplementary advice seems fine to me. I never used it, but I read the forum and Googled a handful of times about specifics I was unsure about. Like pitch accents extending across particles, or the 'ng' variant (forget the technical term ;p). But that was more corroborative.

I'm not *against* adding pronunciation tips w/ audio to beginner materials, since there's already usually phonology stuff (at least for the texts I've used/seen) and I think having audio whether you use it or not is always nice. And seems like it'd be better to slowly introduce pronunciation stuff supplementing examples meant mostly for other things, if you're going to do that.

But as all these threads I'm disagreeing with merge per the conflicted debate we've arrived at as Thora has noted, some of the 'calls' for pronunciation resources seems a bit like asking for advanced prescriptive grammatical explanations to be required up-front.

bounaparte, thanks for explaining, but I think there's been some kind of miscommunication again. The merits of grammar study wasn't even mentioned. A few of us responded specifically to your surprise or outrage that grammar references don't come with audio. That's all.

As Nadia and Nest0r already mentioned, I don't think anyone would consider it a bad thing if someone were to convert grammar references to digital and include audio ... and make it available for free.

But what you are suggesting is that audio for text grammar references ought to be a priority. That's it's necessary. Considering its intended purpose, limitations and the alternatives, I don't agree. If I were an educator or publisher I would instead create audio resources with greater benefit and no limitations.

So, no, it's still not "obvious" to me why you need a good reference grammar with good audio by native speakers. I don't get why you are annoyed by its absence. Best I can tell, you were afraid that reading sentences without audio would ruin your pronunciation at the beginning. If so, I can assure you that you needn't have been so concerned.

Also, no one disputes the importance of pronunciation. We understand that there are several different ways to go about it, that's all. Naturally, we use the ones we think are best, but apparently don't feel compelled to post our surprise when others don't do it our way. The forum would grind to a halt.

So continue to do what works for you, of course, but try not to be so puzzled when other learners, publishers, academics, fellow polyglots, teachers see things differently. It's to be expected when everyone else is a "learned moron", right? :-P

kanjidevourer Wrote:'just approximate the myriad of sounds with this one sound...

I don't think the idea is to just fudge it with one sound. It's more about not needing to apply 6 pronunciation rules when you encounter a new word. It's important to develop an ear and be able to distinguish and produce the different sounds. As your vocabulary increases, you'll be able to read with the correct sound, without thinking about any rules.

Thora Wrote:

kanjidevourer Wrote:'just approximate the myriad of sounds with this one sound...

I don't think the idea is to just fudge it with one sound. It's more about not needing to apply 6 pronunciation rules when you encounter a new word. It's important to develop an ear and be able to distinguish and produce the different sounds. As your vocabulary increases, you'll be able to read with the correct sound, without thinking about any rules.

Can you be sure about this? Have you heard many learners speak? My impression is that the majority of them are terrible, and I've always attributed it to the commonly held "Japanese pronunciation is easy" belief that means few people bother paying attention to anything other than ラ行.

We at least have evidence that some people (including myself) can't accurately reproduce ん simply through listening or even shadowing. I ended up giving it attention after a communication breakdown because of the way I pronounced 全員 more like 善人. Even after making it adequately understandable, my friend pointed out to me that I wasn't extending it for a full モーラ, so words like 全巻 sounded more like two syllables and therefore 訛っている.

The second friend happened to be a native Japanese linguistics 大学院生, and we worked on it during a little pronunciation course she designed as part of her thesis. She didn't cover the sort of lingo magamo used, she focused on what was necessary and approachable, and through a mixture of reading explanations, listening exercises, practice and correction, and even a couple of songs, I was left with a better perspective on things I thought I already knew inside out. If my average native friends couldn't point them out, what chance did I ever have? I was lucky to have that experience, and it's the sort of thing I had in mind when I mentioned a pronunciation course.

nest0r Wrote:

Splatted Wrote:Thx for all the links etc everyone.

nest0r Wrote:I don't see much point in audio designed specifically for pronunciation. Between the many sources of clear audio, ranging from single word sources like JDIC, to smart.fm, to expansive native materials (w/ prosody) enabled via subs2srs...

.......Seriously?

Indeed. I'm surprised that you had to ask whether I'm serious, or are rhetorically asking to express disagreement rather than stating something more straightforward. There's no need for extensive, separate pronunciation courses.

Sounds like you just need more practice listening and speaking based on clear native audio.

It wasn't a rhetorical question. I just wanted to make sure before I went to the effort of trying to convince you otherwise. It seems alot of what I would have said has already been posted by others (especially Javizy), but I'll add to it a little bit.

Even if we it is the case that a person would achieve good pronunciation just from listening to and imitating good audio, surely it would be still be quicker and easier if they had some guidance on how to do it. It's the same as learning a musical instrument; You may be able to tell if you sound rubbish, but working out what to do about itis often far from simple. Also, what is gained from waiting till you realise for yourself that their is problem? My OP is the perfect example of this. Did I gain something from all the time I spent pronouncing けんい as ke/n/i?

Splatted Wrote:

nest0r Wrote:

Splatted Wrote:Thx for all the links etc everyone.


.......Seriously?

Indeed. I'm surprised that you had to ask whether I'm serious, or are rhetorically asking to express disagreement rather than stating something more straightforward. There's no need for extensive, separate pronunciation courses.

Sounds like you just need more practice listening and speaking based on clear native audio.

It wasn't a rhetorical question. I just wanted to make sure before I went to the effort of trying to convince you otherwise. It seems alot of what I would have said has already been posted by others (especially Javizy), but I'll add to it a little bit.

Even if we it is the case that a person would achieve good pronunciation just from listening to and imitating good audio, surely it would be still be quicker and easier if they had some guidance on how to do it. It's the same as learning a musical instrument; You may be able to tell if you sound rubbish, but working out what to do about itis often far from simple. Also, what is gained from waiting till you realise for yourself that their is problem? My OP is the perfect example of this. Did I gain something from all the time I spent pronouncing けんい as ke/n/i?

I think that time wondering whether you should convince me of something you're still new at: you should've spent reading what I and others said more clearly, following your own advice in this comment, ironically enough.

No one here is telling you not to ask for pronunciation advice of the sort you asked above or that all pronunciation advice is bad. That would be silly. But you really can't plan for all those little things people notice and are unsure about when they're beginning. If you want to convince someone to develop and release such materials and you want to waste time following such a course, I can see a handful of people here that agree with you, or would've agreed with you when they were learning Japanese, although I think most people who've begun learning Japanese in this day and age attend to more useful strategies.

The fundamentals of pronunciation aren't something you're going to learn about by reading about nostrilization and bilabial defenestration. That kind of info different learners will need is contingent and variable.

Another tip: Trust your ears. Use the text.

As to pronunciation, grammar and language learning.

Look at this: 'to jest'
Do you know how to pronounce it?
Of course, you must be joking.
OK, say it aloud. I can't hear you. I still can't hear you. Where's the play button? The link is broken. There's no play button.

I've been looking at 'to jest' for twenty minutes and I still don't know how to pronounce it.


My neighbour has a daughter. She can read anything, give her a newspaper, she will read it. Give her a psychology book she will read it. She is extremely clever, but she's only four and still has trouble with some sounds in her mother tongue.

I wrote: 'to jest pszczoła' and asked her to read it. 'To jest pscoła' she said.

I wrote: '発音' and asked if she could read it. She laughed and said it was a picture, you look at pictures, you admire them or abhor them, you can't read them. It's a funny picture, says she, it reminds her of a scary clown.
I said you can read pictures like these.
So how do you say it? she laughed.
'Hacuon' I said.
'Hacuon' she repeated. Why don't you write it the way you say it, then.
I wrote: hatuon.
She looked at it. But it says 'hatuon', you can't even write, she said.
You're right, I said. In fact you write it はつおん or ﾊﾂｵﾝ.
You're pulling my leg, says she. Those are not letters, I can't read them.
I can, I said.
How?
Because I know. I learned how.
How did you learn?
I listened and looked, that's how...

I wrote: 'know'.
'Knof (k-n-o-f)' she read.
No, I said. You say it 'no'. And I wrote 'no'.
You're pulling my leg, she said. You say 'noł' and you write 'know' and 'no'? The same?
Yes, I said. More or less. But I don't say 'noł', I say '/nəʊ/. http://en.wiktionary.org/wiki/know So you know now?
Yes, she said. I know you're pulling my leg.
Which leg? I asked. You have two legs.


By the way, 'To jest pszczoła' means 'This/that is a/the bee'. That what it means. But you still don't know how to pronounce it.

LETTERS DON'T TALK.

So your point is, you can't pronounce Japanese without first learning how to pronounce it. Good point. None of us realized that.

JimmySeal Wrote:So your point is, you can't pronounce Japanese without first learning how to pronounce it. Good point. None of us realized that.

My point is you can't learn how to pronounce ANY language without listening to it.
So if you need to use a grammar book for example, and it is not recorded, you're bound to waste a lot of time, make HELL of a lot of errors as far as pronunciation is concerned.

That's my point, JimmySeal.
By the way, how do you pronounce your nick? And should I pronounce it the English way or the Japanese way? Or perhaps the Russian way?
As to my nick, I like it to be pronounced the Italian way.

buonaparte Wrote:So if you need to use a grammar book for example, and it is not recorded, you're bound to waste a lot of time, make HELL of a lot of errors as far as pronunciation is concerned.

* Unless you have some other source of audio examples and have learned the rudiments of how the language is pronounced.

Problem solved.

My nickname is pronounced the English way, and when I'm reading this forum out loud in front of my computer, I'll be sure to continue pronouncing yours in Italian.

JimmySeal Wrote:

buonaparte Wrote:So if you need to use a grammar book for example, and it is not recorded, you're bound to waste a lot of time, make HELL of a lot of errors as far as pronunciation is concerned.

* Unless you have some other source of audio examples and have learned the rudiments of how the language is pronounced.

Problem solved.

My nickname is pronounced the English way, and when I'm reading this forum out loud in front of my computer, I'll be sure to continue pronouncing yours in Italian.

OK, シールさん (I pronounce your nick in my best English),

here's a simple sentence from a grammar book (no audio):
猛暑のせいか裸で寝るのが癖になった。

Are you sure you would be able to pronounce it correctly at a glance?
With proper sounds, pitch accent, rhythm and intonation?
(Let's suppose you're not fluent in Japanse yet.)

Multiply the sentence by three thousand - that's how many sentences there are in the grammar book.

Why on earth should I look for other sources to know how to pronounce THIS PARTICULAR sentence?
And now imagine you have a play button, you just click et voilà, you hear!

' Wrote:

JimmySeal Wrote:

' Wrote:So if you need to use a grammar book for example, and it is not recorded, you're bound to waste a lot of time, make HELL of a lot of errors as far as pronunciation is concerned.

* Unless you have some other source of audio examples and have learned the rudiments of how the language is pronounced.

Problem solved.

My nickname is pronounced the English way, and when I'm reading this forum out loud in front of my computer, I'll be sure to continue pronouncing yours in Italian.

OK, シールさん (I pronounce your nick in my best English),

here's a simple sentence from a grammar book (no audio):
猛暑のせいか裸で寝るのが癖になった。

Are you sure you would be able to pronounce it correctly at a glance?
With proper sounds, pitch accent, rhythm and intonation?
(Let's suppose you're not fluent in Japanse yet.)

Multiply the sentence by three thousand - that's how many sentences there are in the grammar book.

Why on earth should I look for other sources to know how to pronounce THIS PARTICULAR sentence?
And now imagine you have a play button, you just click et voilà, you hear!

Depending on where you're at in your Japanese study, you'll be able to pronounce it properly with varying degrees of success. Most importantly, your priority in pronouncing it will differ. I don't think grammar sentences should be prioritized as 'I need to pronounce this perfectly'. You don't even need to know the kanji to study grammar. For instance, at the beginning, you can study grammar while still doing RTK, minimizing interest in the phonology as you focus purely on structure of common patterns. This is why books like Japanese the Manga Way have multiple scripts (ローマ字, kana, kanji, etc.) and keep explanations of pitch accents to a minimum.

I think future development of resources should/will continue the flow of dynamic digital multimedia examples with minimal supplementary information that can be customized per user according to target goals.

A compartmentalized production line approach is still the best, IMHO.

I think buonaparte's point is that when you're a beginner (still needing to get a grip on some kind of minimal grammar and basic vocabulary in order to make a smoother translation to native sources), you are reliant on the translations and explanations in grammar references/textbooks/parallel texts audio courses such as pimsleur, teach yourself, jpod101 and such. The audio component is what makes pimsleur etc good but they tend to be extremely limited in their scope. If you want to get a good coverage of the basic grammar necessary to transition to native sources as quickly as possible while also learning a bunch of vocabulary along the way, a decent grammar reference (preferably something going simple to complex with a lot of built in recursion ala heisig) is optimal. The problem is that studying from such grammar books does relatively little for your listening or pronunciation skills. If it was standard practice for audio to be included in such books (and this will likely happen as people move away from paper books) then you could be effortlessly improving you listening comprehension and accent right from the outset while learning basic grammar and vocabulary. This would also allow to transition to native audio sources much faster as you'd recognize the grammar fragments etc. The closest thing widely available now that would allow you to improve grammar/vocabulary and listening is probably parallel texts with audio.

buonaparte Wrote:LETTERS DON'T TALK.

No, people read.
(btw, is there some reason you yell often?)

Quote:My point is you can't learn how to pronounce ANY language without listening to it.

Which doesn't mean that a learner must only read with audio. Saying that listening is important to language learning isn't saying much.

Quote:So if you need to use a grammar book for example, and it is not recorded, you're bound to waste a lot of time, make HELL of a lot of errors as far as pronunciation is concerned.

[Edit: In short, a long-winded detailed version of JimmySeal's concise comment, plus the importance of reading without audio. lol.]

As Jimmy mentioned, by the time learners are reading sentences like those their recognition of sounds and pronunciation ought to be sufficient. The most important period for basic sounds is at the very beginning. There are only a few sounds that English speakers (for eg) have difficulty with. If there are some stubborn problems, they ought to work on those basics. Listening to grammar dictionary sentences isn't a great way to do that and will be ineffective for certain learners.

There are a predictable handful of phonology rules that learners tend to have some difficulty with. There are techniques to work on those, several of which involve no audio. Again, audio for grammar references is far from an ideal solution (especially for for learners who have difficulty distinguishing sounds as is often the case with long and short sounds).

Pitch apparently applies to about 14% of vocab. What most learners need to work on more is how to eliminate intonation transfer from their L1 and improve rhythm. Using audio to correct those problems is just one technique. Non-audio techniques (some of which are the complete opposite of mimicry) are effective and, for some learners, necessary.

Developing reading skills requires some reading without audio. Sustained reading is also necessary, not checking the audio after each sentence. Learners will not ruin their accents by reading without audio. This seems to be the crux of your argument.

I've been around Japanese language learners for *gasp* 25 years. In my day (wheeze...) we worked on pronunciation and pitch at the beginning, but there was little audio available outside of the labs and sessions with TAs. Nevertheless, many of my peers and later students went on to careers in Japanese with decent or great accents. Please stop trying to scare the kids! :-) There's such a fantastic variety of audio available these days (and ways for individuals to manipulate it) that learners have little reason to worry. They also need to develop their inner voice, which means turning the audio off sometimes .

(What I'm hearing, though, is that self-learners might benefit from greater awareness at the beginning of the common problem areas and how to avoid or fix them. That seems like a fairly easy thing for this crowd to remedy.)

Again, if future electronic grammar dictionaries have audio, no one will protest. Nadia has good suggestions for future resources that are better than current audio courses, but without the limitations of grammar references. But to suggest, bounaparte, that learners should have audio for all text all the time isn't good advice, imo. Doing both listening/reading and reading alone is better than only listening/reading. It develops different skills. Folks wanting to work on pronunciation, prosody, vocab, reading, grammar automaticity, listening comprehension and natural speech can continue to use resources better suited to those purposes.

Note to Splatter: there really are so many parallels bw language and music - here, ear training. And the brain scientist are discovering stuff which supports what people have suspected. I find it fascinating, even though I don't really understand it.

Thora,
a very interesting post.

You wrote

Quote:Non-audio techniques (some of which are the complete opposite of mimicry) are effective and, for some learners, necessary.

Would you mind elaborating a little bit, it is something new for me.