Honestly, I can't really note the difference between:

http://en.wikipedia.org/wiki/Voiceless_p..._affricate
VS
http://en.wikipedia.org/wiki/Voiceless_a..._affricate

The second one is the /ch/ sound in (知人/chijin) which is not the same English /ch/ sound like in chop. Before I can even make this sound I need to be able to hear it. I don't think this is a mayor problem to get yourself understood, but it's an accent thing Japanese natives will surely notice right?

Yes, this is something Japanese people will notice -- the pronunciation of "ch" is one of the things that won't be a barrier to understanding but is part of the foreign accent. You don't need to know the definitions of those terms you linked, but if you can pronounce each sound it will improve your accent.

Do you have any tips on how to do that? I am actually busting my head trying to figure this out.

I think t͡ʃ is pronounced with the lips extended, while t͡ɕ is pronounced with them in their neutral position. Just do that and I think everything will fall into place. t͡ɕ has a higher pitch and is softer than t͡ʃ.

Pretty sure the difference is in the tongue positioning. ち is like mandarin q,j and x with the tip of the tongue staying behind the lower front teeth, and the middle of the tongue arching up against the roof of the mouth and blocking the air. English 'ch' is produced by holding the tongue tip against the alveolar ridge (behind the upper teeth, google it) so it's kind of halfway between the mandarin q,j and x and the retroflex sounds ch, zh and sh.

The difference is clearly explained in the Wikipedia article though. But if you want an intuitive answer, like others said, the Japanese version sounds lighter and softer to me too. As you can tell from the fact that these two fall into different categories in phonetics, it's not really hard to hear the difference. It may not mean it's easy to pronounce it correctly though. In any case, it's used for ち, ちゃ, ちゅ, ちょ, and ちぇ. So if you want to focus on this particular sound, you might want to listen to sentences with theses characters carefully.

Also, if you want to perfect finer points in pronunciation, I think you should be really determined. The difference you pointed out is kind of easy one. For example, there are differences which the IPA system can't describe well (Actually Japanese has sounds like う which can't be described by IPA accurately in the first place, but that's slightly off-topic, I guess.). An example of this is how English and Japanese "m"s are different. The former is normally pronounced with your lips touching tighter while the latter is usually softer in this regard and you move to the following vowel quicker. This is a bit too simplistic a description though. Maybe even subtler is the difference between English "b/p" and their Japanese equivalents. The main difference is the timing you start vibrate your vocal cord; you start using your vocal cord milliseconds faster when you pronounce Japanese "p" than the English counterpart while the very first milliseconds of Japanese "b" is like English "p" because your vocal cord starts to vibrate ssssssslightly later than the onset of the consonant.

And if you don't know, the distribution of allophones can be totally different between Japanese and English too. So just because you mastered each representing sound for every phoneme doesn't mean you can get rid of your foreign accent from your articulation. A rather obvious example is that Japanese doesn't change the sound of "t" the way many American English speakers do, e.g., it won't be a softer "d" in any phonological context in Japanese.

There is also a bit complicated matter about allophones. An allophone of an English phoneme can be a stand-alone phoneme in Japanese. For example, "h"s in "hi" and "he" are pronounced differently in English, though native speakers may fail to realize this because everything is subconscious. The latter version of "h" is the same as a phoneme for the ひゃ series, which is different from the consonant of は. What makes it more complicated is that this consonant is also used in ひ, so it is also an allophone of the consonant of the は series. It's a bit similar to the fact that the Japanese "m" can work as both phoneme "m" and an allpphone of "n". By the way, the "h"s in "he" and ひ do fall into the same category in the phonetic sense, but the latter is slightly closer to the "h" in "hi". It may be a bit difficult to become able to hear differences of this level though.

Also, just so you know, pretty much every "equivalent" sound is different between English and Japanese. Falling into the same category in the IPA sense doesn't always mean they're exactly the same. And it seems like some of the subtle differences haven't been documented in linguistics yet. So you can't learn all of these from technical books or academic papers, let alone from textbooks.

By the way, I kind of feel like you don't really need to be able to hear the difference to get the right pronunciation. It's like you're pronouncing them correctly before you know it if you immerse yourself into the language. I think it's pretty much like how native English speakers don't even know they're pronouncing "p"s differently when they say "pin" and "spin." Probably your conscious mind has little to do with how good your pronunciation is (and how good your grammar is for that matter). In fact, some of the subtle differences I wrote in this post are the ones I accidentally noticed when I was explaining Japanese pronunciation to my friends. The rest are mostly the ones I read about somewhere, and I didn't even realize I'd been pronouncing them the way they were explained until I read about them.

Ah, and if you're serious about pronunciation to the extent that you're reading this long post this much, it's not that hard to train your ears to hear the differences. As scientists say, it might get harder and harder to learn to pronounce them correctly as you get older because it's a motor skill. But it doesn't seem like improving listening skills is as difficult.

About whether it is noticeable or not, I can't remember if I was able to hear this particular foreign accent about "ch". But you might want to consider the fact that native speakers aren't that good at noticing subtle differences in pronunciation. As you probably already know, our ears are optimized to only hear the sounds of our native languages so we subconsciously "mishear" sounds, i.e., a slightly accented pronunciation sounds exactly the same as the proper sound because Japanese monolinguals can't hear your foreign sounds just like "r" and "l" sound exactly the same to them. So if your pronunciation is good enough, basically only kids and babies can hear your accent. Watch out multilinguals who speak both Japanese and your native language, though, because they're the guys who pick up on every single quirk in your accent!

Anyway, because I talk with nonnative Japanese/English speakers everyday, I'm pretty sure their nonnative accents are bleeding into my pronunciation. And I kind of like it better that way. I wouldn't say you shouldn't work on your accent. But maybe you don't need to worry too much either. Whatever you do, you end up speaking like people around you anyway.

Magamo is back? With an informative post as always.

*applauds*

@Magamo, that was very informative.

I understand what you say about being 'trained' to hear sounds, but, for example, before I studied Arabic and didn't know the first thing about the language, I could still hear the sounds that were unique to English, I had no idea how they were produced though.

In this case, I am aware they are different but I can't hear the difference at all, they sound exactly the same to me, it's frustrating.

bcrAn Wrote:@Magamo, that was very informative.

I understand what you say about being 'trained' to hear sounds, but, for example, before I studied Arabic and didn't know the first thing about the language, I could still hear the sounds that were unique to English, I had no idea how they were produced though.

In this case, I am aware they are different but I can't hear the difference at all, they sound exactly the same to me, it's frustrating.

Well, one of the points I wanted to make was that probably you don't need to be able to hear the difference to get the correct pronunciation down though... To make it clear, you're pronouncing correct native sounds when speaking your native language even if you can't hear the difference; there are pairs of different sounds in a language which native speakers can't hear the differences but nonnative speakers may find them quite different. It may be counterintuitive, but that's how language works. And another point I wanted to make was like if you're having trouble pronouncing the exact sound, probably it doesn't matter because most likely native Japanese speakers can't hear the difference either just like you. Sorry if my post was unclear.

In any case, I'd say it's not that big of a deal. Like I said, you'll soon be able to hear the difference unless you're learning the language the wrong way. Besides, I vaguely remember the "t" in "tree" etc. sounded exactly like "ch" to me years ago (I'm talking about the ch-ish "t" in many American dialects. I'm not sure if "t" gets closer to ch in other versions of English.). I guess the difference between ch-ish "t" and real "ch" is much larger than the gap between English "ch" and its Japanese equivalent.

[Edit]

Tzadeck Wrote:Magamo is back? With an informative post as always.

*applauds*

Why, thank you!

JimmySeal Wrote:I think t͡ʃ is pronounced with the lips extended, while t͡ɕ is pronounced with them in their neutral position. Just do that and I think everything will fall into place. t͡ɕ has a higher pitch and is softer than t͡ʃ.

English tʃ is indeed usually pronounced with rounded lips, but it has nothing to do with the tongue, and moving the lips certainly won't make the tongue move.

English ʃ is pronounced with the tip of the tongue touching the alveolar ridge (the flat part behind the teeth) just before it goes up sharply onto the palate. The tongue is curled up.

Japanese ɕ is pronounced roughly in the same place, but instead of using the tip of the tongue, an area further back called the blade is used. The tip of the tongue is not used and usually just lays at the bottom of the mouth, not quite touching the bottom teeth. It's not curled.

Unlike what Magamo said, native speakers are VERY good at noticing subtle differences in pronunciation. They are not good at understanding what's wrong, let alone explaining it, but they can spot it right away. So yes, native speakers will notice this sound right away if it's not right. The reverse is also true -- listen to how a Japanese says "Japan" and you can notice right away that something is wrong with the J (and probably both vowels, and the n, and the unaspirated p, for that matter).

Tzadeck Wrote:Magamo is back? With an informative post as always.

*applauds*

Don't leave us again! We need you! Not to exclude any others, but I always find yours and Katsuo's posts to be informative and helpful. Always. It'd be a shame to lose such a valuable forum member as yourself...

AlexandreC Wrote:

JimmySeal Wrote:I think t͡ʃ is pronounced with the lips extended, while t͡ɕ is pronounced with them in their neutral position. Just do that and I think everything will fall into place. t͡ɕ has a higher pitch and is softer than t͡ʃ.

English tʃ is indeed usually pronounced with rounded lips, but it has nothing to do with the tongue, and moving the lips certainly won't make the tongue move.

Saying it has nothing to do with the position of the lips is an exaggeration and saying that moving the lips certainly won't make the tongue move is also false.

I don't know about you, but I can't produce a t͡ʃ with my lips retracted, and I can't produce a t͡ɕ with them pursed. The tongue and roof of the mouth don't meet up to produce those sounds if the lips are out of place, and likewise, positioning the lips correctly will position the rest of the mouth to produce the sounds you want to make.

it always strikes me when i watch j-tv, the difference between 'gaijin with excellent japanese' according to jpnese ppl involved, and those native speakers. they speak very differently, mainly in that the gaijin enunciate much more.

i'm not directing this at OP, but i think perhaps trying to put on a perfect native accent is a bit of a pipe dream for the first 5 years of study... and that a better goal is to speak like 'a gaijin with excellent japanese.'

worth considering is that part of a native accent is not just the correct adoption of nuance, but also the specific corruption of speech that is not actually a positive in the strictest understanding of language study.

Let me take the position of a language learn. It takes time to learn or notice ち. Only dedicated studying on improving ones listening and speaking will this ability improve. That is listening very carefully to each word independently, and in sentences. That would include at different pitches.

One thing that I need to say is to try to concentrate on Japanese pronunciation in the context of the Japanese language. Remember in Japanese the only difference between [ti] and [chi] is the Romanticizing system that uses it. They are both [ち].

In the context of language learning it is great to understand / t͡ʃ / VS / t͡ɕ /. I really enjoyed reading about it.

AlexandreC Wrote:Unlike what Magamo said, native speakers are VERY good at noticing subtle differences in pronunciation. They are not good at understanding what's wrong, let alone explaining it, but they can spot it right away. So yes, native speakers will notice this sound right away if it's not right.

Are you sure? I can't remember if the two "ch"s sounded that difference to my ear, but I just can't convince myself that native speakers are good at noticing subtle differences. I mean, to be honest, I couldn't hear the difference between, for example, ʤ and ʒ used in Japanese, which is my native language... As a native speaker, I've been using them correctly my entire life, but they sounded exactly the same until I learned to hear the difference. Besides, if you listen carefully, there are many native speakers out there who don't follow standard pronunciation rules due to dialects, personal idiosyncrasies, etc., and they don't sound like having a foreign accent at all. Japanese "r" is particularly highly personal in that it's a whole mess of different sounds you can say as "r", and none of them makes you sound very foreign unless your pronunciation is ridiculously off like substituting the typical American "r".

I can understand native speakers would easily notice something is wrong if the different "ch" gets in the way and affects surrounding sounds or the overall flow in his speech. But if you have native speakers listen to the two versions as isolated sounds and ask them which is which, I'm not sure if we can pass a blind test. If it's quite noticeable only in speech, I'm guessing it'd be because it's quite hard or impossible to blend that "ch" naturally so you inevitably screw up the following vowel, interrupt the flow or something along those lones.

Mixing up ʤ and ʒ can be a whole different story though because they're not free allophones; the former is signaling a grammatical boundary. So cases like this may be quite noticeable, though native speakers can't hear the difference if said as isolated sounds. But, ch? Hmm. If I remember correctly, it doesn't have a complicated allophonic rule, so I tend to think it shouldn't matter much. Well, I might be wrong on this one. I'm no expert.

Anyway, I just tried to replace every ち with the English equivalent in a couple Japanese sentences to see if it makes a difference, and I kind of sounded like putting more emotion into it than usual. Also, my English "ch" does extend lips compared with my native Japanese "ch". Reading the description of the voiceless palato-alveolar affricate, I think I can do it either way, but I think what the OP is talking about is the particular version with the lip movement used in English. In fact, if I only change the place of my tongue to get a presudo-English ch (i.e., lips etc., remain the same as the real Japanese one), I can't really say swapping the consonants made a significant difference. My impression was it's well within the range of a normal accent. I might be missing something though because it never occurred to me that it could be without lips extended in the IPA sense. I don't know what your average Japanese would think of the substitution either. I could be subconsciously using more Japanized "ch" than I should when I was doing it too.

JimmySeal Wrote:

AlexandreC Wrote:

JimmySeal Wrote:I think t͡ʃ is pronounced with the lips extended, while t͡ɕ is pronounced with them in their neutral position. Just do that and I think everything will fall into place. t͡ɕ has a higher pitch and is softer than t͡ʃ.

English tʃ is indeed usually pronounced with rounded lips, but it has nothing to do with the tongue, and moving the lips certainly won't make the tongue move.

Saying it has nothing to do with the position of the lips is an exaggeration and saying that moving the lips certainly won't make the tongue move is also false.

I don't know about you, but I can't produce a t͡ʃ with my lips retracted, and I can't produce a t͡ɕ with them pursed. The tongue and roof of the mouth don't meet up to produce those sounds if the lips are out of place, and likewise, positioning the lips correctly will position the rest of the mouth to produce the sounds you want to make.

I can certainly produce tʃ without rounding my lips, and tɕ while rounding them. You are used to the 2 movements happening together, but they are indeed completely disconnected (and this depends on the language). As much as you sound convinced of the opposite, the lips do NOT affect the position of the tongue, and certainly not the position of the palate.

magamo Wrote:I mean, to be honest, I couldn't hear the difference between, for example, ʤ and ʒ used in Japanese, which is my native language... As a native speaker, I've been using them correctly my entire life, but they sounded exactly the same until I learned to hear the difference. [...]

Mixing up ʤ and ʒ can be a whole different story though because they're not free allophones; the former is signaling a grammatical boundary. So cases like this may be quite noticeable, though native speakers can't hear the difference if said as isolated sounds. But, ch? Hmm. If I remember correctly, it doesn't have a complicated allophonic rule, so I tend to think it shouldn't matter much. Well, I might be wrong on this one. I'm no expert.

A native speaker of Japanese will easily recognize if the English ch is used. However, he will not notice ʤ and ʒ in Japanese as the 2 allophones represent a single sound -- just like English speakers won't notice if their vowels are nasalized or not. Perhaps we weren't looking at things from the same perspective.

I'm really surprized to read that you don't find much difference between the English and Japanese ch. I could hear it right away when I started learning Japanese and I hear it in Japanese speakers all the time. I hear in Korean speakers too, and I always notice English speakers speaking Japanese using the wrong one.

As I was teaching pronunciation to a Korean student, she was actually quite surprized to learn that English ch was different from hers. In a way -- as you purport -- she had never noticed the difference or cared about it. But when she finally made the English sound herself, she was extremely impressed to see how different they were and couldn't believe no one had ever told her before. I suppose this would imply that if no one tells you that 2 sounds differ in 2 languages, one would assume that they are 2 allophonic versions of the same sound. Of course, that doesn't make sense if you are comparing distinct languages...

AlexandreC Wrote:A native speaker of Japanese will easily recognize if the English ch is used. However, he will not notice 2 allophones sounding different because in his language, they represent a single sound -- just like English speakers won't notice if their vowels are nasalized or not. Perhaps we weren't looking at things from the same perspective.

Since you seem to know a lot about this kind of stuff, what I'm saying is that /tʃ/ (especially the version without lip rounding) is probably a free allophone of phoneme "ch" (or should I write [tɕ]?) in Japanese just like many seemingly foreign sounds are actually free allophones of [r] in Japanese. This is why I think a native speaker would find it difficult to notice the difference. At least /tʃ/ without rip rounding sounds that way to me. Of course, I wouldn't deny the possibility that my ear got too used to /tʃ/ or nonnative accents in general though.

[Edit] And as you probably already guessed from what you wrote in the edit, most likely English doesn't accept /tɕ/ as a free allophone of "ch". I think that's why your Korean student and I can clearly hear the difference if they're said in English context.

[Edit2] Fixed typos.

magamo Wrote:

AlexandreC Wrote:A native speaker of Japanese will easily recognize if the English ch is used. However, he will not notice 2 allophones sounding different because in his language, they represent a single sound -- just like English speakers won't notice if their vowels are nasalized or not. Perhaps we weren't looking at things from the same perspective.

Since you seem to know a lot about this kind of stuff, what I'm saying is that /tʃ/ (especially the version without lip rounding) is probably a free allophone of phoneme "ch" (or should I write [tɕ]?) in Japanese just like many seemingly foreign sounds are actually free allophones of [r] in Japanese. This is why I think a native speaker would find it difficult to notice the difference. At least /tʃ/ without rip rounding sounds that way to me. Of course, I wouldn't deny the possibility that my ear got too used to /tʃ/ or nonnative accents in general though.

[Edit] And as you probably already guessed from what you wrote in the edit, most likely English doesn't accept /tɕ/ as a free allophone of "ch". I think that's why your Korean student and I can clearly hear the difference if they're said in English context.

[Edit2] Fixed typos.

Allophones are variations (predictable or not) used by native speakers of a language and which are mentally represented as a single sound. As far as I know, tʃ is not used in Japanese -- at least, I've never heard it in Standard or Tokyo Japanese -- so it's NOT an allophone. The fact that native speakers would readily perceive [tʃ] as meaning [tɕ] doesn't mean it's an allophone. It's simply a bad pronunciation caused by L1 interference.

AlexandreC Wrote:Allophones are variations (predictable or not) used by native speakers of a language and which are mentally represented as a single sound. As far as I know, tʃ is not used in Japanese -- at least, I've never heard it in Standard or Tokyo Japanese -- so it's NOT an allophone. The fact that native speakers would readily perceive [tʃ] as meaning [tɕ] doesn't mean it's an allophone. It's simply a bad pronunciation caused by L1 interference.

I think some native speakers including me sometimes use /tʃ/ as in the initial "ch" of「チェッ！　上手くいったと思ったのに。」. The closer it gets to English /tʃ/, the more emotional it sounds. Also, I feel it tends to shift to /tʃ/ more often when it's followed by vowel う, especially when the speaker is putting stronger emotion.

Also, I know Wikipedia isn't a reliable source, but the Japanese page of [tʃ] says:
日本語の「ち」や「ちゃ」行の子音は、話者によって異なるが、標準的な発音は 無声歯茎硬口蓋破擦音 [ʨ] のほうである。
i.e., Japanese Wikipedia says some speakers do use /tʃ/, albeit they're not the majority. This definitely matches my intuition.

Are you sure you never heard /tʃ/ used by native Japanese speakers in natural conversation? I think I hard it many many times in my life.

Is this a very subtle difference or is it something completely obvious? I still can't note the difference.

Are both articulated with the blade of the tongue? Is the position of the lips the same in both phonemes? I am not particularly psyched about having a perfect Japanese accent, but I am studying IPA thoroughly and I just can't get around this.

As for this video (

) , is that a joke? I hear the same thing. Indeed, there is a difference in that one is a bit more soft than the other, but is that enough to be classified as a different phoneme?

I hear a very subtle difference... the second 'ch' sounds more 'juh' ish

I still don't understand how to produce it.

I see many people posted indicating it is softer, the tongue is placed against at the alveolar ridge, etc., but that is just like the English ch. I mean, what do you do (or not do) in the Japanese /ch/ compared to the English /ch/.

bcrAn Wrote:Is this a very subtle difference or is it something completely obvious? I still can't note the difference.

Are both articulated with the blade of the tongue? Is the position of the lips the same in both phonemes? I am not particularly psyched about having a perfect Japanese accent, but I am studying IPA thoroughly and I just can't get around this.

As for this video (

) , is that a joke? I hear the same thing. Indeed, there is a difference in that one is a bit more soft than the other, but is that enough to be classified as a different phoneme?

The first one (i.e., /tʃ/) is used in English. But as far as I know, native English speakers round their lips a little, which, as I understand it, isn't a necessary condition to be classified as the sound in IPA phonetic sense. But if you don't do it, the sound gets softer and lighter. In other words, there are at least two versions within /tʃ/ which Japanese-English bilinguals might make a distinction of. Since English isn't my native language, I have no idea how the average native English speaker hears them. Anyway, I believe this non-lip-rounded version is used by some native Japanese speakers as an alternative sound. The Japanese Wikipedia article seems to be supporting this as well. Because they're allophonic in Japanese, which means they represent the same phoneme in Japanese, your average Japanese would find them quite similar.

It seems the second one (i.e., /ʨ/) isn't used in English, and when used in Japanese, your lips don't get rounded. So at least in our context, this has only one version.

The difference between the English /tʃ/ (with your lips rounded) and the Japanese /ʨ/ shouldn't be extremely subtle to people who speak both English and Japanese to an extent. So it's quite unlikely you fail to notice as long as your learning method is reasonable.

The only difference is the place of articulation in the IPA sense. So lip rounding shouldn't matter in this sense. But the actual realization of /tʃ/ in English requires lip rounding while that of /ʨ/ in Japanese doesn't. So if you're talking about phonemes in English and Japanese, yes, lip rounding is also a distinctive factor. If anything, I think your average monolingual Japanese would find it quite hard to hear the difference between /tʃ/ and /ʨ/ if both are without lip rounding.

bcrAn Wrote:I still don't understand how to produce it.

I see many people posted indicating it is softer, the tongue is placed against at the alveolar ridge, etc., but that is just like the English ch. I mean, what do you do (or not do) in the Japanese /ch/ compared to the English /ch/.

Read the Wikipedia articles you yourself linked to (especially the line that reads, "Its place of articulation is..." in each one). If English isn't your native language, you might find it helpful to click your native language on the left side if available and go to the article in the language.