Talk:Romanization of Japanese/Archive 1

Archive 1

Archive 2

Archive 3

"Romanji"

hmm. I thought it was Romanji -Sv

Nope. It appears to be a fairly common misspelling, though. --Brion 00:49 Oct 19, 2002 (UTC)

Corrected. -- Bugbread 12:12 Jan 17, 2004 (JST)

The long vowel mark cannot be displayed under windows' Japanese encoding. user:Ktsquare

Hmm, I'm running (US) Windows 2000; it comes up fine in Mozilla 1.2beta and Internet Explorer 5.5. What are you running? --Brion

It looks fine now. Kt2

I've very frequently seen rōmaji referred to in English; I've virtually never seen it misspelled "romanji". I am therefore changing the text accordingly. -- Hoary 08:32, 2005 Mar 13 (UTC)

It might be worth adding a note on this particular misspelling somewhere in the article, as it is very common. I've just come from Hiragana where an anonymous editor actually "corrected" romaji to romanji. It would be nice to have an explanation to point such people to. — Haeleth Talk 14:22, 24 May 2006 (UTC)

Preferred romanization

Question:
ISO 3602 (Kunreisiki, Nipponsiki) is the valid system nationally and internationally. And The US system (Modified Hepburn) was abolished in 1994. Wikipedia should not oppose the standardization.

1. That's a statement, not a question. 2. Wikipedia will not abandon Hepburn, as that is used worldwide. Kunrei is frequently used in Japan, but is not used that much elsewhere. However, I am inserting the appropriate Kunrei in several articles, e.g. Shoujo. WhisperToMe 03:37, 11 Apr 2004 (UTC)

Romaji is not only used when transcribing japanese for foreigners, it is part of the japanese language. It is often used as short for words. As an example, "OL" is a very common word which is short for "Office Lady".

Kunrei-shiki standardizes romaji as part of the Japanese language. The English language Wikipedia has articles on people, places and things as they are known and can be pronounced in English -- which is generally not by kunrei-shiki. I don't think that using Hepburn opposes the Japanese government's standardization at all, since that standardization is internal and refers to the Japanese language, not English. -- Tlotoxl 05:24, 23 May 2004 (UTC)

The Hepburn-system is linguistically invalid; no Japanese linguist supports it. The might of the Hepburn system is oppsed to linguistics; it is politically held by US imperialism. I strongly oppose any usage of the Hepburn system. I have never encountered a single Japanese linguist who supports that invalid system. Abolish it now.Everton 18:10, 16 Jan 2005 (UTC)

This has been fought over before (again and again and again). Hepburn is not "invalid", quite the contrary it's a lot better for non-linguist English speakers than any of the alternatives, and that's Wikipedia's main audience. Jpatokal 05:38, 17 Jan 2005 (UTC)

Many linguists explain why Hepburn is invalid. The most apparent invalidity of Hepburn is that different letters are used for one phoneme. (sa shi su se so, ta chi tsu te to) That destroys the phonemic structure of Japanese and make confusion. By using the Hepburn system, Japanese grammar becomes far more difficult and complicated.

Those who support Hepburn really do not know Japanese language. I do not think that the problem should be determined by those uninformed people.Everton 20:56, 17 Jan 2005 (UTC)

One word: "Fujitsu". -- Curps 22:24, 17 Jan 2005 (UTC)

"Fujitsu" is completely wrong even in the Hepburn system, because it lacks the long sound mark on the final letter. It should be "Huzitû"

Without long sound mark, Japanese cannot be read correctly. "tôri" (street) and "tori" (bird) cannot be distinguished. "Yôko", "yoko", "yokô" are also completely different words. People who write "Fujitsu" or "Tokyo" are completely indifferent of correctness. I think those invalid writings must be banned. Everton 22:59, 17 Jan 2005 (UTC)

The company itself calls itself "Fujitsu", are you suggesting that we "ban" (whatever tbat means...) Fujitsu? And if you look at Fujitsu, you'll see that the article notes that the name is a contraction of Fuji Tsūshinki Seizō.

Also, "yoko", "yōko", and "yokō" are different in Hepburn too. Macrons can and should be used in content for any Japanese long vowels. Jpatokal 12:20, 18 Jan 2005 (UTC)

First, the existence of the Japanese word ōeru and its "OL" etymology and (even amid Japanese script) orthography tell us nothing whatever about the status of Hepburn.

Secondly, I wholeheartedly agree that Hepburn is an abomination. True, few Japanese linguists use it -- in their examples and discussion. But try looking at the way they write their own names on those very books or articles (if in English): usually, they employ Hepburn. Moreover, Kunreisiki isn't up to the demands of 20th-century Japanese phonology.

But most importantly, Wikipedia is not a soapbox. However ghastly Hepburn is, and however deplorable its use is, it's very widely used and requires a good explanation. -- Hoary 08:32, 2005 Mar 13 (UTC)

I couldn't agree more -- which is why I've removed the Kunrei spelling you just added. See Wikipedia talk:Manual of Style (Japan-related articles) if you, or somebody else, wants to beat dead horses some more. Jpatokal 09:26, 13 Mar 2005 (UTC)

If I'm going to beat something, it might as well be a dead horse. Er, I mean, I hadn't even heard of the existence of that talk page. I'll digest it in due time, and till I've done so shall beat other things.

As long as we're not shouting at each other, I'll add a comment. I came to this page to see what it said about romanization: I believe (or, if you prefer, have deluded myself) that I'm fairly well informed about it, and was wondering whether W'pedia had good information that I could recommend to a Japanese-non-speaking friend. The reason? I had to explain why I was, with extreme reluctance, using Hepburn for a certain (Wikimedia-unrelated) project. And the reason was that I had to transcribe gairaigo for which Kunreisiki doesn't cater. The article on Kunreisiki talks of adding apostrophes, but I have never seen this done in the real world, and use of apostrophes would seem to require explanation and anyway seems an awful kludge. Pity, because I hate to use Hepburn. -- Hoary 14:20, 2005 Mar 13 (UTC)

Everton writes above that The Hepburn-system is linguistically invalid; no Japanese linguist supports it. ... I have never encountered a single Japanese linguist who supports that invalid system. I'm not entirely sure what this means. Of course the huge majority of Japanese linguists have little or no interest in one system or another; it's merely a tool for an expository job. If on the other hand there's an implication that no Japanese linguist uses Hepburn, this is plain wrong. A counterexample is Tsuyoshi Ono and Sandra A. Thompson, "Japanese (w)atashi/ore/boku 'I': They're not just pronouns" (Cognitive Linguistics 14 (2003)), which De Gruyter has kindly placed on the web for all to see. (I happened to come across it just now while googling for something rather different.) -- Hoary 04:04, 2005 Mar 16 (UTC)

I don't think anyone wants a pokemon called "pikatyū" 201.239.182.69 22:36, 9 October 2006 (UTC)

JSL

Is JSL actually commonly used? I have an earlier book here by Eleanor Harz Jordan (Beginning Japanese, Part 1, 1963) in which she uses a different ad hoc system of romanization, called "BJ romanization". Was JSL ever taken up by any significant number of other texts? I don't think we need to give a whole table-column and listing to a romanization that is only used in five or so books anywhere. Compare the rare transcription mode in which small 'tsu' is 'q' -- e.g., 'niqpon': I've seen that used in various linguistics texts by different authors, although admittedly never in public use. --Aponar Kestrel 19:44, 2004 Jul 27 (UTC)

The only place I've ever seen JSL is in textbooks from the 60s too. Exploding Boy 01:21, Jul 28, 2004 (UTC)

Ronmaji?

Must confess, I've never heard of rōnmaji. Can anyone confirm this, or is it a typo for the incorrect romanji?

It was a typo, I think I've already corrected it.

Question

Now I'm really really confused. Which system (hepburn,kunrei-shiki,nihon-shiki) should I used in Japan? --202.99.60.155 03:20, 15 Oct 2004

Many people will understand Hepburn thanks to their English training. For the same reason, Nihon-shiki is probably not a good idea, although it does make the most sense from their perspective, and would be understood. Kunrei-shiki walks the best balance and is probably best understood by the most people, but Hepburn is probably second choice after that, and I'd say is the most common to be seen on products where they have professionals working on it. Really, it doesn't matter; they all represent the same thing, and to a large degree it's a matter of personal preference. Most of the people I talk with online don't stick to any one system, they just mix and match, because it doesn't make a difference. Bigpeteb 13:39, 14 Oct 2004 (UTC)

Hepburn is the most common, hands down. Exploding Boy 15:28, Oct 14, 2004 (UTC)

In the English-speaking world, perhaps. In Japan, it is rather mixed. In Elementary schools, students still learn nihon-shiki/kunrei-shiki during the 4th grade as part of their Kokugo (Japanese) classes. Although I am told that they then learn a variant of hepburn in their English classes, this only really leads to more confusion. The primary usage of romaji for Japanese is for Japanese-character input on computers, and in that particular context, the hepburn system is inferior to Nihon-shiki and Kunreishiki. Of course, most input method systems are very flexible and allow mixing and matching, but for typing speed, hepburn is likely to be the slowest. Zubari 15:47, 2004 Oct 30 (UTC)

The less you use rōmaji in Japan, the better (INMHO) --81.60.3.240 10:22, 22 November 2005 (UTC)

Splitting the article

Although this doesn't need to be done right away, I think this eventually needs to be split into two articles:

• Romaji, dealing with the use of the Latin alphabet in the Japanese written language
• Japanese romanization, dealing with the transliteration of Japanese into English

Most rômaji encountered by native Japanese people isn't transliterated; it appears in text alongside kanji and kana. Things like "OL" and "H suru" and "W hai" and "Y sha" and so on. A completely different subject matter than transliterating text for foreigners, and it doesn't generally involve the Hepburn/kunrei debate at all.

At any rate, people wanting to know how Japanese is romanized will probably not know the word rômaji. - Sekicho 14:42, Nov 1, 2004 (UTC)

There is an article Transcribing_English_to_Japanese, which you may or may not be aware of. I think we should have an article like Japanese writing system, similar to Chinese writing system. As you notice, English alphabets are intergred parts of Japanese writing system. It seems to make sense to have one central article overviewing the writing system and if the topic grows too much, separate it. -- Taku 17:09, Nov 1, 2004 (UTC)

I agree... the writing of English in Japanese (Namely the Alphabet letters) doesn't have anything to do with romanization. I wonder how that ever made it on this page o_O -- SuperMidget 15:48, 10 May 2006 (UTC)

Modern Systems Consistent Spelling

Through the discussion of the Modern systems of transliteration Nippon-Siki is listed as a main form. However in the Kunrei-shiki section this is then spelt Nihon-Shiki. Which is correct, also should it be Shiki or Siki? This really should be standardised throughout this text as it's quite confusing, however I don't know which is correct so cannot change it. Ben W Bell 15:22, 8 Mar 2005 (UTC)

Both: Nihon-shiki is correct in Hepburn, but Nippon-siki is correct in 日本式 itself. But Wikipedia prefers Hepburn, so Nihon-shiki it is. Jpatokal 17:05, 8 Mar 2005 (UTC)

Cheers. It doesn't really matter that much which one, but this article should only have one spelling in it, the nuances of the differences can be left in the Nihon-shiki article itself. Ben W Bell 20:03, 8 Mar 2005 (UTC)

Article title: Romaji or Rōmaji?

Now that the article title has been renamed from Romaji to Rōmaji, let's retrospectively discuss if this was a good idea. Over at Everything2.com, I remember it was possible to have titles with Unicode characters by using character entity references, however the general policy did not approve of this. I checked the manual of style for Japanese articles and it says article titles "must use short vowels" and that apostrophes are too confusing. One thing to note is that circumflexes were possible to use in article titles even before the switch to UTF-8 encoding, but weren't used. However macrons are more popular and now any Unicode character including letters with macrons can be technically used. I don't understand how avoiding apostrophe can actually reduce confusion instead of increasing it, but oh well, that's the policy. —Tokek 17:13, 13 July 2005 (UTC)

I'll just leave a link: Wikipedia talk:Manual of Style for Japan-related articles/Romanization where there has already been a somewhat cluttered discussion on this. —Tokek 02:59, 24 July 2005 (UTC)

The current discussion on this topic is located at Wikipedia talk:Manual of Style (Japan-related articles)#English Wikipedia is now UTF-8—Tokek 16:06, 31 July 2005 (UTC)

Addition to Non-standard romanization

I think it might be worthwhile to mention common mis-uses of romanization, commonly used by foreigners and Japanese alike.

While obviously people are standing by the ti, tu, si, and hu romanizations, others are pretty much universally shunned. (I hope.)

For example:

• sya/syu/syo instead of sha/shu/sho or even shya/shya/shyo
• ah instead of ā/â/u (this point is debatable I guess)
• uh instead of ū/û/u (rare, but understandable given the above and 'oh')

And even quite oddly (by non English speakers) you can often see long vowel sounds with 'assumed' consonants for example:

• バーガー (burger) romanized as berger/burgur (proper style would produce bāgā or baagaa)
• スロープ (slope) romanized as slorp
• コート (coat, or coart) romanized as coult/coalt

and many more examples. While these are not strictly romanizations (more like attempts at spelling English by non-native speakers) it may be useful to note some incorrect uses to clear up some confusion that people feel when trying to understand signs that are supposedly readable by English speakers or by people able to read English characters. Gavin 2005 07 15

sya/syu/syo: those are used in Kunrei-shiki & Nihon-shiki romanizations which are not non-standard. Spelling coat as coart, etc. seem to be just English spelling mistakes. I don't think listing incorrect permutations is appropriate for an encyclopedia. The Engrish article already makes fun of English spelling mistakes by Japanese people, I don't see why we need yet another. —Tokek 16:44, 15 July 2005 (UTC)

JSL

Why were all references to JSL removed from this page. It may not be in common usage, but I don't think that we need only discuss common romaji systems here. If it's significant enough to warrant its own article, it certainly needs at the very least a passing mention in the article on romaji. Nohat 23:10, 28 August 2005 (UTC)

Romaji and Language learning

I would appreciate seeing a good article on the factual basis (if any) for the often cited claim that romaji impairs one's learning of Japanese. Are there any empirical studies that have looked into this? Any studies that have actualy compared classes of students whose method of instruction differs only in the use of romaji versus kana? What lead to the widespread change from romaji-based texts to kana-based texts? Was it just intellectual fashion, was it based on empirical findings, was it driven by marketing of texts as being written in "real Japanese"? Does anyone know?

I've always thought this claim was unlikely to be based on real evidence. The series of books I used, Japanese for Busy People, had kana from the beginning, but it is hard to understand (a) how anyone could measure the difference (b) why romaji could possibly make such a difference, since kana do not actually have any pronunciation information in them. Generally speaking, language learning (not just Japanese) is notorious for having fashionable "wonder" methodologies claimed to affect students, but for which it is extremely hard to measure real differences. English language learning has probably hundreds of them, and they change like fashions. There is no real scientific basis for any of them. The biggest differences in speed of learning certainly aren't caused by using romaji or kana, etc., but by factors such as student motivation. Any claim that romaji actually affects foreign students negatively is probably nothing more than someone's opinion. --DannyWilde 05:05, 15 September 2005 (UTC)

Circumflex on Nihon, Kunrei

Kunrei and Nihonshiki use circumflexes, not macrons. Take a look at the link to romaji soudan shitsu, which has details from the standards documents. In fact I think they used macrons for Nihon shiki then switched over to circumflexes. Anyway the standard says circumflexes. --DannyWilde 07:35, 11 September 2005 (UTC)

The page's Modified Hepburn isn't such?

This article claims that long vowels in Modified Hepburn are represented with vowels with macrons. However, according to the definition in the Hepburn romanization article, it's actually Revised Hepburn that does this, whereas Modified uses the doubled-vowel notation. That article says that "Revised Hepburn may be referred to as modified Hepburn" (and vice versa) but I think things should be uniform for clarity's sake. Thus, should we change the references in this article to "Modified Hepburn" to read "Revised Hepburn" instead? -- J44xm 23:52, 14 September 2005 (UTC)

You seem to have checked it carefully so I suggest that you go ahead and change it as above. --DannyWilde 05:07, 15 September 2005 (UTC)

Quick check on this one

For example, there is no accepted way of romanizing the common combination 'ウィ' of katakana u and small i, used in such words as ウィークリーマンション (literally, "weekly mansion", i.e. an apartment rented by the week). The romanization systems are sometimes informally extended, so that for example ウィ is sometimes romanized as "wi".

Can I get a sanity check on this? I tried searching Google and found exactly one case of "ウィークリー" being romanized with "wi": [1]. Also, usually, wouldn't this be romanized as "uiikurii"? Anyway, the "tu" example is OK - I didn't see any rationale for changing it, it's probably the clearest example of any, since the "フィ" etc. ones get changed to "fi". Any comments appreciated. --DannyWilde 12:31, 18 October 2005 (UTC)

• Good point. What happened was that I thought of my own real-world problems in romanizing ウィ and then thought of a common word that used it. I lazily didn't check if that romanization was common. I'm amazed to see now that Google estimates about twice as many hits for トゥ as for ウィ: I'd thought that the former was very peripheral indeed, and the latter commoner. Meanwhile, the trouble with "uiikurii" is that it implies ウイークリー (big イ). -- Hoary 13:38, 18 October 2005 (UTC)

DannyWilde's edits and Wapuro romaji

While I think DannyWilde has done generally a good job of whipping the article into shake, some parts of the article — notably the massive list of nonstandard romanizations — now overlap heavily with wapuro romaji, which is not even linked anymore. I'd suggest retaining the status quo — discuss the big romanization systems here, and the bizarre stuff incl. 'unromanizable kana' (a concept I disagree with, as they're diacritics, not standalone characters) in there. Jpatokal 16:55, 18 October 2005 (UTC)

That's a good point, wapuro romaji should not have been unlinked. It's difficult to keep track of that stuff, so thanks for spotting that. I didn't intend to unlink it. --DannyWilde 03:16, 19 October 2005 (UTC)

The list of nonstandard romanizations is interesting, but it certainly is rather long and I wonder whether it's encyclopedic. (People complain about fancruft on Wikipedia; could this be labeled rōmajicruft?)

Don't be ridiculous. --DannyWilde 03:16, 19 October 2005 (UTC)

As for 'unromanizable kana', I disagree with the concept myself if romanization is, as I'd thought, more often than not the representation of Japanese sounds (phonemes, of course, not phones) in roman letters, rather than the representation of Japanese writing in roman letters. Not 'unromanizable kana', then, but 'unromanizable phoneme combinations' -- though the latter should preferably expressed in a less rebarbative way. (But of course one of the problems with the real-world romanization of Japanese is that different systems have different understandings even of this: consider how the postpositions は、を、へ are usually written wa, o, e [representing sounds] but in some [quasi-] systems instead as ha, wo, he [representing graphemes].) Whatever we call this section, it is worthwhile, as real-world Japanese has phoneme-combinations that just aren't covered even by "modified" (and remodified etc.) Hepburn; when fixes/kludges exist at all, they may or may not coincide with those of FEP/IME rōmaji input. Hoary 21:01, 18 October 2005 (UTC)

PS I've removed this section: [QUOTE]

On a computer or word processor, these smaller kana may be produced in various ways. For example, an 'x' or an 'l' preceding the roman input for a kana produces a small version on some systems, thus xtu gives 'っ' on many IMEs. However this is not standardized, and these "spellings" are restricted to use in input systems; they are not used to represent the smaller kana in romanized Japanese.

[UNQUOTE] -- which seems a bit of a red herring. If the first 7/8 or so belongs anywhere, it is indeed to an article on input. -- Hoary 01:51, 19 October 2005 (UTC)

There isn't any sensible article on input, so you'd better make one, and put that stuff in there, before editing the current article like that. The word IMEs is wrong, the abbreviation IME of "input method editor" is the name of a Microsoft product, in fact. --DannyWilde 03:16, 19 October 2005 (UTC)

Please don't throw stones in glass houses — it really would have been wise to discuss what you had in mind before rewriting half the article. Jpatokal 12:15, 19 October 2005 (UTC)

What's your objection to the article Wāpuro rōmaji? (How is it not sensible?) -- Hoary 11:42, 19 October 2005 (UTC)

Further, there is an article on Input method editor, suggesting to me that it is, or anyway can be, a generic term. Just as "FEP" is an obvious abbreviation of "front end processor", "IME" is an obvious abbreviation of "input method editor". Before I decided that this paragraph anyway didn't belong here, I'd wanted to write "FEPs", but decided that "IMEs" could be more familiar to many. Has MS registered "IME" as a trademark? Anyway, the IMEs of both my Mac and my SuSE box use "x" in the same way as the 'Doze IME does. -- Hoary 11:51, 19 October 2005 (UTC)

add kanji and English to all examples

moved from Talk:Romaji – ABCD^e✉ 00:46, 15 February 2006 (UTC)

Add kanji and English to every place it is possible to do so.

E.g., what are the kanji for "Nihon-shiki". also say ("Japan style") if that is indeed what it literally means. Don't make me click to hopefully find out. I'm reading this offline. Nippo jisho too. JSL too. What is that in English even, too?

E.g., "Yajiro" certainly has kanji.

Say what a macron is. Add an example of JSL.

What does "and [[]], and [[]], and [[]], and [[]]" mean to the average reader?

"Wapuro is a contraction..." well, mention that at the FIRST occurrence of wapuro.

"Since the war": say which war.

Maybe mention why the same character is sometimes read kawa, sometime gawa (river.)

210.200.105.231 20:27, 9 February 2006 (UTC)

Romanization of "N"

Wouldn't エン be a better romanization of the letter "N"? (originally unsigned by Louisng114, 2006-04-05 16:12:41 EST

I agree that it would, but I think エヌ/enu is the official transliteration. ん/ン didn't exist when kana was created, and the Japanese probably transliterated the English alphabet and the letter N before ん/ン was created. So, they just kept it as「エヌ」。

- KevinJr42 09:22-25, 23 April 2006 (UTC)

There's an "official" transliteration? (And I'm pretty sure ん has been around for a good bit longer than English has been known in Japan.) More likely it's due to the fact that if N were エン, acronyms like NBC エンビーシー would sound like MBC instead. Regardless, I've fixed the spelling of NHK. --Aponar Kestrel (talk) 23:20, 6 May 2006 (UTC)

(And then promptly bungled it again, when adding the citation. Thank you, Kusunose.) --Aponar Kestrel (talk) 07:26, 7 May 2006 (UTC)

Uh, yeah... Kana is definitely older than the knowledge of English in Japan. I think n is the only kana without an inherent vowel sound, mainly because there were so many Chinese loanwords ending in -n. 惑乱 分からん 09:43, 14 May 2006 (UTC)