3

Are there written systems (preferably international & standardized) that you can use to depict how to pronounce any word from Japanese and any language. Something similar to Romaji but one that can be used for any language?

Can't remember well but back in a linguistics class, I thought they showed me some kind of international alphabet (which just looked like modified Roman alphabets with weird tildes, lines, and such) that you can use to break down and identify any particular sound and pronunciation for any particular language. They may have called it a phonetic alphabet or something.

Example

So, what I mean is if I want to show a Japanese person how to properly pronounce "加油", I can turn to this international written system. The Japanese person, in turn, can use the same phonetic notation to show how to pronounce "Ganbatte".

What I want

  1. Does such a phonetic notation exist?
  2. If so, can you provide a list mapping the sounds these characters make using that phonetic system?

    あ,い,う,え,お,か,き,く,け,こ,さ,し,す,せ,そ,た,ち,つ,て,と,な,に,ぬ,ね,の,は,ひ,ふ,へ,ほ,ま,み,む,め,も,や,ゆ,よ,ら,り‌​,る,れ,ろ,わ,ゐ,ゑ,を

    Think on Wikipedia calls them the Hiragana base characters.

    No Kanji though.

I hope that makes sense.

9
  • Does kana itself not fulfill that role? They tell you exactly how to pronounce the sound, and if you show someone a kanji with the kana (ルビー), they'll know exactly how to read/pronounce the kanji. Maybe I'm missing something here...
    – istrasci
    Nov 9, 2012 at 16:22
  • @istrasci Yes, but looking for something can be used for any language as well, not just Asian languages. Or, imagine if communicating with ppl who don't know English. Need some international standard we can refer to
    – damx
    Nov 9, 2012 at 16:27
  • @istrasci 加油 is a Chinese word meaning 頑張って. The poster wants a way to express the phonetics of the Chinese word. Kana will not help.
    – Dono
    Nov 9, 2012 at 16:29
  • @istrasci Yes, what he said. Sorry, if I am not expressing myself well. It's out of what I know.
    – damx
    Nov 9, 2012 at 16:32
  • OK, I think I understand better now. Sort of like an international ルビー, right?
    – istrasci
    Nov 9, 2012 at 16:32

2 Answers 2

10

1)Yes, an international standardized character alphabet exists for transcribing the sounds of all human Languages. It's called the International Phonetic Alphabet and it is maintained by the International Phonetic Association (both are acronymized as IPA). The most recent version of the alphabet was created 1969 and their most recent and currently operative handbook, first published in 1999, is the Handbook of the International Phonetic Association: A Guide to the Use of the International Phonetic Alphabet. Although as a guide, it's rather useless as I explain below, and certainly is not for beginners. If you need to learn the IPA, pick up a good introductory linguistics textbook, it should teach you the IPA and any background info you need to understand the alphabet, like articulatory nomenclature. If you aren't already familiar with baisc linguistics terminology, then the Handbook will be a very difficult read.

2) There is a concept called orthographic depth which is a measure of how well a language's orthographic symbols are in 1-1 association with the language's phonemes. Phonemes are the finite discrete combinatorial units of sound that are unique to a given language, though all languages draw their phonemic inventory from essentially the same superset of phonemes since all humans have the same speech organs (tongue, larynx, alveolar ridge, etc.) and it is our speech organs which define what sounds are even possible. The entity being represented and transcribed by an IPA symbol is a phoneme. It is somewhat a curious thing how every language, including Japanese, possesses a highly intricate and constantly changing set of rules that govern how these phonemes can combine to produce the verbal form of words. This set of rules describing how the phonemes pattern to form the words of a given language is called a phonology for that language. Languages with the same inventory of phonemes can have different phonologies ie different rules in how those phonemes are combined to produce words. IPA symbols representing phonemes are concatenated and placed in between two '/' characters to represent the phonemic representation of words. So, <今日は> is phonemically transcribed as /koNːit͡ɕiɰa/ using the IPA. A sequence of phonemic characters is called a string, a term borrowed from computer science.

However, even though there is a standardized alphabet, the transcription for a given phoneme, sound, or word can differ depending on how you use the alphabet. The IPA is only an alphabet, it doesn't tell you how to interpret and transcribe what you hear. It is still left to devise a scheme for associating the IPA symbol to the phoneme you are hearing. And there is usually several mutually contradictory schemes that you could define for transcribing the phonemes of a given language's phonemic inventory that are all consist with the IPA. It usually comes down to how you segment the speech stream. When you study phonetics, you'll know what I mean. If you look in the Handbook at the languages they exemplified, you'll notice it's referred to as an "illustration of the IPA for language X". This is because there are different ways to transcribe with the IPA. For example, sometimes you can add diacritic marks to make the transcription more accurate, but only if you really need that extra detail.

The IPA is only for transcribing phonemes, it says nothing about how a language is to combine those phonemes to form words. In other words the IPA is agnostic to any particular phonological theory.

Most kana are in 1-1 correspondence with pairs of phonemes, but several kana are 1-1 with singular phonemes. So, <む> corresponds to the phoneme pair /mɯ/ but <あ> corresponds to the phoneme /a/. <ん>, however, is a little more complicated to describe. It correlates with an underspecified phoneme, a special type of phoneme that the IPA doesn't accommodate because it's a language dependent phonological conception. There's no room to explain that here. Because most kana are in 1-1 correspondence with a single or pair of phonemes, kana is said to have a shallow orthographic depth, but it isn't perfect 100% 1-1 correspondence when considering the entire Japanese script and it also depends on how you define orthographic depth. These technical definitions are important if you are trying to do something like getting a computer program to read Japanese text and synthesize speech. Such a task is called grapheme to phoneme (G2P) conversion.

Kanji on the other hand is more or less arbitrarily associated with phonemic strings. In other words, when you encounter a novel kanji for the first time, you cannot predict its phonemic correlate ie how it is pronounced. Because of this Kanji is said to have a deep orthography.

You can type out IPA symbols then copy and paste them however you like using this website http://westonruter.github.com/ipa-chart/keyboard/ . This is how I type IPA to include in web pages. You need to set your browser to view UTF character encoding to see IPA characters. Do that if the IPA characters are not rendering.

Since kana is orthographically shallow you can construct a list that maps kana to IPA phoneme strings rather easily. But you need to establish your transcription conventions beforehand, and to what level of detail you need to transcribe. For these reasons, a phoneme to kana implementation is usually created only for the application at hand.

4
  • 1
    I wrote a sloppy machine readable mapping a little while back, you can look at it to just get a sense of the size and complexity of the task forum.gaijinpot.com/…
    – taylor
    Nov 10, 2012 at 5:38
  • 2
    "<ん>, however, has several different phonemic correlates". Are you sure of this? Do you have an example minimal pair? I always thought of ん as having one phonemic correlate /N/, but many phonetic realizations, [m][n] etc.
    – dainichi
    Nov 11, 2012 at 0:41
  • @dainichi yeah, you're right, it's underspecified. I need to think about how to fix that without mentioning allophones though.
    – taylor
    Nov 11, 2012 at 0:49
  • I should point out that in my answer here I only mentioned phonemic string writing with the IPA. Phonetic string writing, which is the transcription of allophones not phonemes, takes place with the same symbols but '[]' braces are used instead of '//' slashes as delimiters. Phonemes are preverbal "units of sound", they don't actually have an acoustic signature. A phoneme string is retrieved, processed and converted into a allophonic string (which has a phonetic signature) during speech articulation.
    – taylor
    Nov 11, 2012 at 1:03
7

This is what you're looking for: International Phonetic Alphabet

9
  • I think that's it. Is it possible you can map the kana sounds using that system, which is ultimately what I would like. I myself don't know how to use it.
    – damx
    Nov 9, 2012 at 16:29
  • @damx It could be done, but it will be imprecise at best. Kana is a spelling system, not a phonetic alphabet. It is not precise enough for a complete phonetic mapping of Japanese. Treatment of long vowels, for example, may be problematic. Also some sounds such as ɴ/m/ŋ (ん). Also, there is no way to express suprasegmental aspects such as accent. This could become very lengthy, so I will leave it at this. But let me add one more link: en.wikipedia.org/wiki/Help:IPA_for_Japanese
    – Dono
    Nov 9, 2012 at 16:37
  • 1
    Well, I just want the sounds the "hiragana base characters" make. The small chart here en.wikipedia.org/wiki/Hiragana#Writing_system "あ,い,う,え,お,か,き,く,け,こ,さ,し,す,せ,そ,た,ち,つ,て,と,な,に,ぬ,ね,の,は,ひ,ふ,へ,ほ,ま,み,む,め,も,や,ゆ,よ,ら,り,る,れ,ろ,わ,ゐ,ゑ,を" It's around 47. Not really concerned too much with ん anyway or long vowels. Just those listed really.
    – damx
    Nov 9, 2012 at 16:50
  • @damx Wiktionary has IPA representation for each of the hiragana characters. ex.
    – ento
    Nov 9, 2012 at 17:22
  • 1
    The problem with this is that the kana don't always have the same sound. The IPA is meant to describe phonetics, not as a transliteration table between scripts. For instance, ん has as much as FOUR different phonetic representations in IPA, depending on how it is pronounced.
    – vivien
    Nov 9, 2012 at 21:39

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .