Romanized Japanese

Doug putting the foot down and pushing for Romanization?

The problem with Japanese (and East Asian languages as a whole) is that inference of the context is immensely important, especially in conversation. Some words could have entirely identical readings with different meanings altogether, and if not for the different kanji writing, might as well be indistinguishable. Moreover, written Japanese has three scripts, all of which can and should be used in conjunction. Hiragana for particles and words, Kanji for compound words and local names, and Katakana for loan words and foreign names. Romanizing all that isn't impossible, but it won't really make learning the language much easier, AFAIK.
 
Some words could have entirely identical readings with different meanings altogether, and if not for the different kanji writing, might as well be indistinguishable.
Korean has the same issue (worse than Chinese or Vietnamese, actually, due to lack of tones differentiating eg 韩 and 汉) and yet Hangul is perfectly understandable even without using Hanja.
 
Korean has the same issue (worse than Chinese or Vietnamese, actually, due to lack of tones differentiating eg 韩 and 汉) and yet Hangul is perfectly understandable even without using Hanja.
How so? Is it mainly by infering the context, or something I missed. I'm not familiar with Korean, sorry. :coldsweat:
 
How so? Is it mainly by infering the context, or something I missed. I'm not familiar with Korean, sorry. :coldsweat:
By context.

If homophones X and Y can be distinguished in speech by context (as homophones are in most cases, except in intentional conflations like puns), they must logically also be distinguishable in a phonetic approximation of speech (i.e. writing) by context.

This is why you can write a book solely with pinyin and have Chinese people understand it.
 
By context.

If homophones X and Y can be distinguished in speech by context (as homophones are in most cases, except in intentional conflations like puns), they must logically also be distinguishable in a phonetic approximation of speech (i.e. writing) by context.

This is why you can write a book solely with pinyin and have Chinese people understand it.

But that's... what... I... said... ._.

EDIT: Never mind, I understand. Basically, it's possible to Romanize the script and still have a working language (basically applying how you understand the conversation by context into the written word).
 
Last edited:
Off topic, but Romanizing Japanese is probably one of the most un-Japanese things to do.:frown:
 
This excerpt from Tae Kim's Guide to Learning Japanese gives one of the main disadvantages to a potential Romaji-only writing system:

Why Kanji?

Some people may think that the system of using separate, discrete symbols instead of a sensible alphabet is overly complicated. In fact, it might not have been a good idea to adopt Chinese into Japanese since both languages are fundamentally different in many ways. But the purpose of this guide is not to debate how the language should work but to explain why you must learn Kanji in order to learn Japanese. And by this, I mean more than just saying, "That's how it's done so get over it!".

You may wonder why Japanese didn't switched from Chinese to romaji to do away with having to memorize so many characters. In fact, Korea adopted their own alphabet for Korean to greatly simplify their written language with great success. So why shouldn't it work for Japanese? I think anyone who has learned Japanese for a while can easily see why it won't work. At any one time, when you convert typed Hiragana into Kanji, you are presented with almost always at least two choices (two homophones) and sometimes even up to ten. (Try typing "kikan"). The limited number of set sounds in Japanese makes it hard to avoid homophones. Compare this to the Korean alphabet which has 14 consonants and 10 vowels. Any of the consonants can be matched to any of the vowels giving 140 sounds. In addition, a third and sometimes even fourth consonant can be attached to create a single letter. This gives over 1960 sounds that can be created theoretically. (The number of sounds that are actually used is actually much less but it's still much larger than Japanese.)

Since you want to read at a much faster rate than you talk, you need some visual cues to instantly tell you what each word is. You can use the shape of words in English to blaze through text because most words have different shapes. Try this little exercise: Hi, enve thgouh all teh wrods aer seplled icorrenctly, can you sltil udsternand me?" Korean does this too because it has enough characters to make words with distinct and different shapes. However, because the visual cues are not distinct as Kanji, spaces needed to be added to remove ambiguities. (This presents another problem of when and where to set spaces.)

With Kanji, we don't have to worry about spaces and much of the problem of homophones is mostly resolved. Without Kanji, even if spaces were to be added, the ambiguities and lack of visual cues would make Japanese text much more difficult to read.

That said, claims of homophone problems in any language are almost always overblown, since any given pair or set of homophones will usually occur in different syntactic contexts (e.g. no English speaker, upon hearing "please open that can of tuna" is going to be confused and think "can" here means "be able to").
 
Last edited:
Nihon-shiki or not, this is unfeasible due to many homonyms. I have started learning Japanese a couple months ago and getting the correct character when typing on a computer (you type the romaji and you get the character is the basic method) is a problem because you can get so many varied characters for the same romaji string.
 
As much as it pains me to say it, it's not unfeasible if you just get rid of all the words that came from Chinese, which is where I'd wager 98% of the homonyms are to be found. For example "gyo" is a Japanese reading of a bunch of Chinese characters that can have meanings as varied as "fish," "jade," "dawn," etc. Japanese has plenty of native words for these concepts. Of course, for the Japanese themselves it'd be the equivalent of making English speakers jettison Latin and Greek words. ITTL I suppose they would have to rely on English academic terms to an even greater extent.

As a real-life example, Vietnamese got rid of Chinese entirely and while I understand they have some Sino-Vietnamese words, the bulk of the language gets carried on in native vocabulary.
 
Top