Wikipedia talk:Manual of Style/Japan-related articles/Macrons (November 2006)

From Wikipedia, the free encyclopedia
Jump to: navigation, search

What the hell is with all those "macrons" in Japanese names?

It's so ridiculous to see so many Japanese names on Wikipedia have macrons in them when the Japanese people themselves rarely use macrons. The Japanese use either Japanese hiragana/katakana/kanji or English spelling. They just don't use romaji with macrons or whatever except on some rather rare occasions. Do the official documents of the Japanese government use macrons for Japanese names? No. They use either English or Japanese. They don't have a romaji version of text for begining Japanese learners. Do the Japanese person's names on Japanese passport use macrons? No. They are written in English spelling. Only non-Japanese students of Japanese language who learn romaji first thing in their lesson think romaji is all that and using romaji spelling over English spelling is somehow more genuine to Japanese language when it's not. (I'm a native Japanese speaker btw) So please get rid of this ridiculous rule for Christ's sake: "If none of the above is available, use the macronned form." Use it on some English otaku site, not on Wikipedia. --Saintjust 01:05, 12 November 2006 (UTC)

Sounds good (but not really). Which of the dozen or so standards do you suggest we adopt instead of the standard we have adopted up to now? Rhialto 04:03, 12 November 2006 (UTC)
While I disagree with the delivery above, I have to say I too am against using macrons as the default. 1) The macron is rarely if ever used in English (we're not talking about the grave, acute or circumflex accents here, which do have some historical basis). 2) While in principle Choshu and Saigo Takamori might differ from Chōshū and Saigō Takamori, in a well structured article that employs the kanji or kana, I doubt macrons convey much extra information. 3) As a matter of taste, they're just plain ugly, especially when there are a bunch together. I would prefer that we either stick to single "o" or "u," or in a worst case, employ "ou," "oo," "uu" in strictly transliterating the kana renderings. Macrons should at most methinks be relegated to the first instances of words/names or in the nihongo template. I don't go around pronouncing Julius Caesar as "Yoolius Kaisar" (like classicists think the Romans did; nor do I write it Iulius); why bother with the macrons here? The consensus in favor of macrons seems fairly new, so I doubt it will be overturned. I simply wanted to register my thoughts.--Monocrat 04:45, 12 November 2006 (UTC)
The macrons are included as a pronunciation guide to non-Japanese speakers. Academic texts (which I think Wikipedia should strive to be) use macroned versions. Using them is professional, in my humble opinion. Bobo12345 08:59, 12 November 2006 (UTC)
In my opinion article titles are not the appropriate place to teach the pronunciation of a word. Also there is a Wikipedia guideline stating that the most commonly used English term takes precedence over other terms for a subject. Jecowa 09:49, 12 November 2006 (UTC)
It's not just about pronunciation of the word. It's about proper spelling. "Ningyo" (人魚, "mermaid") is not the same as "ningyō" (人形, "doll"), and "ryu" (りゅ) is not a word at all, while "ryū" (りゅう) is, with several possible meanings. The best and most compelling argument I've come up with relating this to the Anglicization of Western languages is the Spanish tilde. "Pinata" and "Cumpleanos" are not words, and, in being spelled wrong, strongly imply or encourage the wrong pronunciation. Meanwhile, "piñata" and "cumpleaños" are words, and reflect the proper pronunciation and spelling in their original language. Again, this isn't directly primarily about teaching others to pronounce it; it's about accurately reflecting the proper spelling, even if it is in a different writing system. Speaking for myself, as a student of Japanese studies, I find it extremely useful to look at a romanized word and be able to figure out or at least guess what its kanji, and thus its meaning or etymology, might be. Things that are spelled "chū" or "chuu" are fairly likely to involve the kanji 中, and to have some sort of meaning related to "middle", while those spelled "chu" are either mispelled or are guaranteed to not use the 中 kanji. As for macrons versus "ou", "uu", I really have no preference. But correct spelling, proper systematic romanization, is essential to any semi-scholarly work; we shouldn't be simplifying the spelling down to what looks simpler and more "English" just for the sake of the least common denominator. LordAmeth 13:45, 12 November 2006 (UTC)
Use of macrons or any standard in transliterating depends on the field, the editing staff, and the author. I'm coming from a purely jingoistic and superficial approach: what looks best to my Anglophone eye. Macrons, especially en masse, spectacularly fail that "test." (Moreover, I find them irritating as a student of both Latin and Japanese.) We have the ability to employ the kanji/kana or refer to articles where they are employed, obviating most if not all of the urgency behind exact transliteration in all instances of a word. I don't see why we can't lessen our burden and employ the forms that are most widely used in English-language press and popular literature: yes, I feel the least-common denominator should be given priority over unsightly and (in this instance) mostly uninformative standard forms. Again, should I render Julius as Iulius or Ivlivs or shift my pronunciation in everyday speech? Pinata vs. piñata: Context is useful in such instances, as I believe it would be in this instance. You've put in a lot of work, Ameth, and I respect that. I'm not suggesting it all be undone straight-away. I'm just saying I disagree strongly with you. :)--Monocrat 17:11, 12 November 2006 (UTC)
You seem to forget that the Wikipedia is an encyclopedia. If this was a huge group blog, doing without the macrons would be okay (although on some days that is exactly what the Wikipedia reads like). Every scholarly work I have refers to Matsuo Bashō. So the fact that a Google test will show that there are about 1,590,000 websites that refer to a macronless "Basho" (compared to the approx. 30,400 that do use the macron) should make absolutely no difference on how the Wikipedia names the article. BlankVerse 17:29, 12 November 2006 (UTC)
(Breaking indent.) An encyclopedia still strikes me as an inherently generalist publication that targets the literate generalist. Were this something called "The Encyclopedia Nipponica" or the "Oxford Dictionary of Japanese Culture and History," I might support the use of macrons. But it is not, so I do not. I'm not calling for the abolition of the macron, I would prefer simply relegating it to the Nihongo template or when absolutely necessary otherwise. Again, I don't see the difference in principle in the indiscriminate use of Bashō and Iulius Caesar.--Monocrat 20:16, 12 November 2006 (UTC)
The obvious difference between those two examples is that I might well read Iulius incorrectly as "Eee-uu-lee-uss", whereas even if I did not understand what a macron meant I would still read Bashō as "Basho". Removing macrons from words is information loss for those readers who can benefit from their inclusion. Bobo12345 22:52, 12 November 2006 (UTC)
This information does not have to be lost. It could be stored in the first line as pronunciation information. Using macrons could also cause pronunciation confusion for people unaware that macrons are used differently for Japanese words than they are for English words. This could result in readers thinking, for example, that "sābisu" is pronounced "Say-bee-Sue." I had never heard of macrons being used in the Japanese manner until I saw them on Wikipedia article. It was very confusing for me. Jecowa 23:21, 12 November 2006 (UTC)
I fail to see how the macron cannot be just as disruptive as the consonantal I. In either case a non-specialist will ignore the notation and attempt to approximate the pronunciation. At least a small annoyance. (As Jecowa noted above.) Apart perhaps from initial instances to ensure comprehensiveness, it conveys 1) no useful information to the generalist (the vast majority of our potential readership), 2) little extra information in isolation to the specialist, and 3) absolutely no extra information when coupled with kanji/kana. Why bother with the eyesore, the complication, or the hassle of converting all the articles?--Monocrat 23:24, 12 November 2006 (UTC)

Essentially, the use of macrons is but one specific component of the romanization scheme that has previously been agreed on. Personally, I think it is an adequate system for describing the information (assuming appropriate page redirects to facilitate searches are in the wiki). It gives full information, and isn't really ambiguous. Yes, it is different from that long vowel marker some dictionaries used, as noted by Jecowa. But just as there are people who have seen that dictionary long vowel marker usage first and would be confused by the romanized Japanese usage, the reverse also applies. Many, many, many words, symbols, phrases, and terms have dual usage. That is no reason to stop using them within their relevant specialised contexts.

Basically, I'm open to change, but I'd like to see a convincing reason to use a specific other romanization scheme instead of the present one with its macrons. Despite a previous person's unfamiliarity with the usage in romanizaed Japanese text, it is a standard method, one of several. Changing the romanization method for the sake of change is bad. Give me a convincing reason first. Rhialto 02:22, 13 November 2006 (UTC)

I am unsure I can satisfy your request for a convincing reason. All I can say is that I propose the change, I suppose it's come to that, for the sake not of change but that of style. I'm indifferent to the scheme chosen to give the rigorous phonetic rendering in English with the Nihongo template or at the outset of an article. A modicum of necessary macrons per article is reasonable. In regular prose, however, I'm really bothered by numerous macrons, and even by double vowels. Wherever possible, I'd much rather collapse to single vowels: Choshu instead of Choushuu/Chooshuu/Chōshū. It's purely a matter of preference--as I have always maintained--but the first is far more pleasing to my eye, and seems less likely to distract or trouble a general reader. We're not all linguists or Japanists. We will have discharged our duty to substance by providing kanji/kana and a rigorous romanization at the outset. Could we then nod to style and adopt the practice of fine anglophone news and literary publications?--Monocrat 03:25, 13 November 2006 (UTC)

Arbitrary division 1

First off, let's be civil. The original post in this thread contains a number of ad hominem attacks. (i.e. Only otaku use macrons, only people who don't know the language use them.) Those arguments are both untrue and invalid to this argument in general. I find entering macrons annoying. (I prefer using WAAPURO-shiki.) I hate having to look in the little box below to find the macron o and hope that I haven's inadvertantly clicked on the Hungarian double accented o or something. The wikiprogrammers could make out lives a bit easier by giving us a drop down box to choose which language's accent marks we want to use. They also make creating redirects more difficult--seiyū names with macrons require four redirects instead of one. However, I also belive in Stare decisis. This had been decided before. Changing the rules like we change our linnens creates a lot of confusion. I think that there is a firm and constant standard which stands on a solid historical and linguistic backgrounds. There's nothing inherently wrong with the current guideline, just a preference in style. We shouldn't make any hasty changes without a good reason. Without a compelling reason to change, it should stay put. --Kunzite 04:59, 13 November 2006 (UTC)

Saying that it was decided before hides the fact that many of the macron decisions were made quite recently (and contentiously). There was a compelling reason given not to use the macron in all situations, and that was how it hides Wikipedia pages when people use search engines. The issue was never resolved to my satisfaction. I don't wonder where all these voices against the macron were then - it is that many people don't watch this page, but have noticed all of their Japan-related articles being changed over the past month. Dekimasu 05:13, 13 November 2006 (UTC)
That reason doesn't strike me as a powerful one. First, Google does pretty well at finding (for example) "choshu": a few seconds ago, Choshu (i.e. "Chōshū (Redirected from Choshu)") came fourth out of a list of an estimated 90,600 hits. Secondly, even if macron use did confuse search engines I wouldn't be so worried, as I think WP is so well known by now it doesn't need search engines so very much. Thirdly, I don't think that the title or content of an article should be altered to suit search engines unless there's good reason. Incidentally, my own main interest in Japan these days is its photography and the best single work in English I know of about the matter is The History of Japanese Photography (2003, ISBN 0-300-09925-8), which though published by a university press is not an academic work but yet provides macrons wherever appropriate and also puts people's names in the right order, rather than the WP practice of reversing the names of those born after 1868. -- Hoary 06:33, 13 November 2006 (UTC)

Most major English-language news media don't use macrons (including the English editions of major Japanese newspapers). Most major English-language encyclopaedias and dictionaries don't use macrons. --61.198.211.5 09:03, 13 November 2006 (UTC)

It's certainly true that most of the former don't. But of course WP isn't a news medium. As for the latter, I really don't know. Perhaps you're right. Rather over a century ago, the compositors of that dodgy but justly popular reference work Things Japanese managed to set it up using the Hepburn romanization of the time, complete with macrons wherever needed (and with no exceptions, thus "Tōkyō" etc); I wonder why doing this is so distasteful a full century later. The only newish reference book with lots of romanized Japanese in it that I happen to have with me now is Bocking's A Popular Dictionary of Shinto (sic), which also consistently uses macrons (thus "Shintō", Tōkyō"). The English-language Japan-related book I've bought most recently is Shōmei Tōmatsu: Skin of the Nation: name inverted for some silly reason but macrons retained (other than on the cover of the book and in writing about it, perhaps in order not to alarm librarians and others). -- Hoary 09:24, 13 November 2006 (UTC)
Ultimately, this is what I think it comes down to: I consider myself a proper scholar, and I think most of us here on Wikipedia ought to take that attitude. What we do here may be a very enjoyable hobby, but it's not a game. The goal here is to create an accurate, comprehensive, and useful encyclopedia. Mispelling things or misrepresenting them to make it more digestible to the lowest common denominator of Anglophone society is not the goal here. I don't see why there needs to be any difference between a "generalist" text and one that's accurate to the subject presented. While I certainly understand your point, "Julius Caesar" is a poor example, as he's so well known by that Anglicized name in English. Take a look at Æthelred of Mercia, Jan Milíč, Curaçao... editors working in other regional topics seem to have no qualms about spelling things correctly. "Choshu" is as much as word as "Jooleeus Seezar" is. You're simply misspelling it to make it easier to pronounce or easier on your eyes or whatever; it's still wrong. LordAmeth 09:35, 13 November 2006 (UTC)
I think Wikipedia should have a scholarly attitude in its approach to quality of research and sources, but there is and should be a sharp difference between scholarly and generalist works. Scholars rightly use a different language to communicate amongst themselves. But Wikipedia is here for the masses, scholars included. And no, Chōshū is not a proper spelling because it is not widely used in Japanese. To my knowledge, romaji is not a native or fully naturalized script, whereas diacrits are are fully part of the script in Central/Easter European languages. If the Japanese were to consistently adopt macrons, I would probably defer. But I doubt they have or will. Setting that aside: Julius Caesar, is an eminently viable comparison: I'm breaking conventional English to closer approximate both the sound and appearance of the original. If that doesn't fly for Romans, why does it fly for Japanese? And why should we not appeal to the lowest common denominator when we have discharged our duty to substance by providing a rigorous romanization at the outset? Your phobia of the LCD is perplexing. While maintaining its standards for quality of research, ought Wikipedia not be accessible, and indeed appealing, to the broadest literate audience? I doubt yielding on the macron issue will lead to our becoming the Springerpedia.--Monocrat 13:47, 13 November 2006 (UTC)

Articles about Japanese topics are likely to also use other words in rōmaji apart from the page title. It's much better to consistently use macrons instead of cropping all words to short vowels and wrong pronunciations or overuse the nihongo template. Yet worse would be to use macrons for the first instace and then drop them – it would look unprofessional and confusing. Sometimes people don't read articles from the beginning.

Macrons actually do give extra information for the casual reader. They can prompt a curious person to learn more about the Japanese language and pronunciation. If a word has a widely known and used form in English, e.g. Tokyo or Julius Caesar, it is already preferred in the current Wikipedia style guidelines.

If you are really annoyed by how macrons look, you might alter your monobook.js to convert all letters with macrons to ones without them. (With "oo" "ou" this would be impossible.) In my opinion, essence is far more important than style, especially in an academic and well researched publication that Wikipedia should be. Any mirror is free to filter the articles into a more eye (or lowbrow?) friendly style. Have you noticed how ugly the reference indicators are? Yet they are essential. Wipe 09:49, 13 November 2006 (UTC)

The monobook suggestion is a good idea, and I'll look into it. However, I'm not opposed to macrons in all instances. It seems that the nihongo template pops or could up in most Japan articles, if only to provide the kanji/kana original, so inserting the macronned form there seems reasonable and sufficient from a scholarly perspective. That is what I mean by providing the macronned form at the outset. Reference indicators are annoying, and essential. Macrons are simply not necessary in every instance of a word. Having high standards doesn't mean we have to make things a chore for the uninitiated.--Monocrat 13:47, 13 November 2006 (UTC)
"Macrons are simply not necessary in every instance of a word." It may not be your intention, but I interpret this to mean that "it is not necessary in every instance to spell a word correctly". Two quick examples that I often see used incorrectly include "to / too / two" and "there / their / they're". If you do not distinguish them in every instance, it is simply wrong. Same with macrons. That does not mean that spelling mistakes do not happen on Wikipedia. If an editor finds correct spelling "a chore", they are always free to spell it however they wish. As with all spelling mistakes, others will correct it. Bendono 14:09, 13 November 2006 (UTC)
This argument has been played to death and I am still much more against the "macron it because it's right" argument than the "let's macron things" proposal/decision. As has been mentioned ad nauseum, the spelling argument only holds if you view English to have an infinite amount of possible letters. If you want to be consistent and say it has 26 letters, you don't use the macron. ワープロ romanization doesn't result in information loss. Dekimasu 04:13, 14 November 2006 (UTC)
English does not have 26 letters. English does not even have an alphabet. The most common script employed to write English is a heavily modified version of the Latin script. There are upper and lower case variations of letters. Those alone exceed 26 letters. Historically you will also find thorn, eth, yogh, ash and other letters. Macrons and other diacritics are heavily used in writing Old English. I gave this list before: English words with diacritics. I count a great deal more than 26 letters. The 26-letter English world simply does not exist. Bendono 06:34, 14 November 2006 (UTC)
The list is populated by words which haven't been fully integrated into the English language. When they are fully integrated they lose their diacritics. Many of the words on the list are just as common without diacritics as with them; when used with diacritics, many of the words on the list are italicized to highlight the fact that they aren't English. My dictionary has 26 sections. Uppercase and lowercase is a shell argument. If you want to call it 52 letters, so be it. A macron is not correct, it is a choice. Dekimasu 07:04, 14 November 2006 (UTC)
I must disagree with your opinion those words are not fully integrated into English. Words are used within a context. When that context is English, those words are just as much English as any other words. Spelling too is a choice. Try this sentence: "To day eye went two too stores and bought to packages of meet witch I will cook to knight." Do you approve of this spelling? Does "correcting" it help you interpret it? Professional, scholarly, academic or not, does this pass here on Wikipedia? And why? This is just as unacceptable for the same reasons as missing macrons. I do not mind people submitting such text. However, I would like the opportunity to fix it. Same with macrons. Bendono 07:44, 14 November 2006 (UTC)
As Rhialto mentioned, there are many different standards for romanizing Japanese words. Macrons are not required to correctly transliterate Japanese names, although they are required under the current guideline. Jecowa 16:12, 13 November 2006 (UTC)
Language, including spelling, is governed by the conventional usage of its speakers and writers. So, you would be right if there were not numerous standards or if there were one of either: a definitive convention of usage among the Japanese themselves, or in lieu thereof an official ordinance from the Government of Japan for a definitive romanization. Please point to either of those to establish what is correct. In fact, from the links provided in Hepburn romanization, the Japanese Land and Foreign Ministries seemingly collapse long vowels to single O and U. Anyone know what the Education Ministry says?--Monocrat 16:35, 13 November 2006 (UTC)
If you really want an official government ordinance then it would be Kunrei-siki. Notice the differences especially in the t- and s- sounds. Long vowels are distinguished (with a circumflex). Bendono 21:30, 13 November 2006 (UTC)
There are a lot of different standards one could stick to, and I'd be happy with most of them. But we need to pick a standard and stick to it. I am soooo tired of arguing this. The point is not to push some kind of macronization agenda. The point is to push an agenda of standardization and consistency. Do you have a problem with standardization and consistency? LordAmeth 09:16, 14 November 2006 (UTC)
I hope that response is not targeted directly at me. I think you and I pretty much agree on romanization and macrons. And I fully agree with picking a standard and being consistent. Monocrat asked for "an official ordinance from the Government of Japan" and that would be Kunrei-siki. I am satisfied with Hepburn (which distinguishes between short and long vowels). Bendono 11:37, 14 November 2006 (UTC)
No, no. I'm sorry. That was not directed at you, that was directed at Jecowa (who I have nothing against personally, for the record). LordAmeth 18:15, 14 November 2006 (UTC)

I just noticed that the article on Riki Choshu (長州力) is now at "Riki Chōshū." Chōshū! LOL. Tell me this is a bad joke.... --61.198.211.5 10:18, 13 November 2006 (UTC)

I'd never heard of this person by any name or with any spelling. So why are you LOL? One possibility I can think of is the absurdity of (a) bothering with the precision and/or pedantry (if that's how you see it) of using macrons while (b) making the elementary gaffe of putting the name back to front (after all, he's not 力長州 but 長州力). Is this it? -- Hoary 10:29, 13 November 2006 (UTC)
Choshu is a very famous pro-wrestler in Japan. What's funny is that every freaking Japanese person is given a strange macroned name just because some ignorant gaikokujn doesn't know about him/her. Romaji fundamentalism could be really funny sometimes..... Seriously, if you have no idea about something, just leave its name alone. --61.198.211.5 12:43, 13 November 2006 (UTC)
So what do you have against being funny, then? These things are always negotiable. Wipe 13:02, 13 November 2006 (UTC)
I don't follow what you're saying at all. Not every freaking Japanese person whose name has 長音 is given a macronned name, strange or otherwise; e.g. I called 十文字美信 plain "Jumonji" as this is how he consistently romanizes his own name. Ignorant gaikokujin though I am, I had nothing to do with the renaming of Riki Chōshū; what makes you think that the person who did it was ignorant of either the man or the pronunciation of 長州? But if what you call "romaji fundamentalism" is "really funny" and has you "LOL", then that's all to the good, is it not? -- Hoary 14:05, 13 November 2006 (UTC)
I know who he is. I am not a fan, but he's on a lot of variety TV shows in Japan. Whether you like it or not, his (stage) name is Chōshū Riki. The macrons correctly reflect how the name is spelled in Japanese. For me, that is the end of the debate and I am satisfied. However, I did not change it. (You can find out who by looking at the history page.) As per the guidelines here, if you feel he is well known outside of Japan in English by another spelling, then you can make that argument and probably get the name changed. I suggest discussing it on the Talk:Riki_Chōshū page where others familiar with him can decide. Bendono 14:23, 13 November 2006 (UTC)
I thought that macrons couldn't indicate Japanese spelling in cases of "ō" which could represent「おう」or「おお」. Isn't that right? Jecowa 16:12, 13 November 2006 (UTC)
That, in fact, is the only drawback to using macrons, and as other have indicated, those who care about it will be able to figure it out pretty easily. And for the macrons themselves, they conveny the appropriate information to those who care about it, and those who don't will simply ignore them. It's a happy medium. ···日本穣? · Talk to Nihonjoe 17:14, 13 November 2006 (UTC)
I thought that "a happy medium" was why macrons were being used in the article text, but not in the article title, per the brokered agreement that you wrote this spring. Dekimasu 04:13, 14 November 2006 (UTC)
Since then, discussion has determined that there is no logical reason to not use them in the titles of articles. As long as proper redirects are created (and in most cases they are), then finding them is not an issue. ···日本穣? · Talk to Nihonjoe 19:15, 14 November 2006 (UTC)
I apologize if you didn't like my changes to Riki Choshu articles, but I don't appreciate the intimation that I don't know what I'm talking about. I may not know much about wrestling, but I definitely know that 長州 is "ちょうしゅう" and not "ちょしゅ". Do whatever the hell you want with these wrestling articles; I'm tired of arguing with you people. Just please don't accuse me of not understanding the Japanese language, history, or culture. LordAmeth 09:16, 14 November 2006 (UTC)

Arbitrary division 2

There seems to me to be an eminently logical reason to avoid macrons: they're officially sanctioned for at most use in the Japanese rail system. Thank you, Bendono, for providing me the information about Kunrei-siki. I have thought on the matter, and it seems only just that if we defer to the French Academy (and other such institutes) to determine the spelling and rules of their languages, we must defer to the Cabinet of Japan. The Cabinet obviously prefer Kunrei-siki (and I ever so slightly prefer the circumflex to the macron), so that system leads the pack, and people truly worried about spelling should prefer it. (I note it's also the ISO standard.) The Cabinet, however, has presumably also approved Passport Hepburn (and possibly also Railway Hepburn). Deferring to the Cabinet leaves us with either the general way of romanizing Japnese or the the manner meant for foreign consumption and comprehension of Japanese. In lieu of a popular convention on the matter, deference to the Cabinet seems the best way to go. If we want to "spell" Japanese in the strictest sense possible, Kunrei-siki; if we want to "spell" it correctly but make it accessible to the whole world--as the Foreign Ministry does on its website[[1]]--Passport Hepburn is the way. In any case, in terms of article prose, macrons seem to me to belong at most in rail-related articles.--Monocrat 01:56, 15 November 2006 (UTC)
Changing to the Kunreishiki system is such a bad idea it's not even worth contemplating. Just because it is officially sanctioned does not mean that it is worthy of use on Wikipedia. God knows why they the Japanese government insists on teaching it in schools. Any speaker of English attempting to pronounce a romanised Japanese word for the first time will pronounce "Tu" incorrectly as "Too" and "Si" as "See" every time. Hepburn romanisation is used widely in academic texts (not only on the railways) because it allows native English speakers to best approximate Japanese pronunciation. It's the best system and we should stick with it. Bobo12345 08:53, 15 November 2006 (UTC)
When "they" teach Kunreisiki in "their" schools, I don't suppose "they" are primarily thinking of its immediate benefit to "any speaker of English". But really, the pros and (considerable) cons of Kunreisiki and Hepburn are so well known and already so laboriously described on WP that all this surely doesn't have to be regurgitated here. -- Hoary 10:44, 15 November 2006 (UTC)
Sorry, when I said "they" I meant the Japanese government and Mombusho, who I don't happen to hold in very high regard. And shouldn't Mombusho be primarily thinking of "speakers of English", since surely that's what they want the students to become? Anyway, I agree this argument is not productive to the project, so let's not bother going round in circles. Bobo12345 12:48, 15 November 2006 (UTC)
I see no reason why Monbusho should pay any particular attention to the needs of native speakers of English when deciding which romanization scheme to employ when teacher English to native speakers of Japanese. Equally, I see no reason to pay any attention to a policy set by an organisation dedicated to the needs of native speakers of Japanese, when our audience are primarily native speakers of English. Rhialto 13:17, 15 November 2006 (UTC)
Teaching English and teaching romanization are separate issues, but for what it's worth, English students are taught to write their names neither with macrons nor in kunreishiki... that is, in the standard English form. Dekimasu 14:05, 15 November 2006 (UTC)
I repeat my question. How is anything created for a Japanese audience relevant to anything created for an Anglophone audience? There's no real causality involved here, just coincidence, and so no reason to use practices in one as justification for practices in the other. Rhialto 15:18, 15 November 2006 (UTC)
(Breaking indent) If we're truly interested in "spelling" Japanese correctly, as Ameth and I believe Bendono have stated, then Kunrei-siki is the way to go. As has been noted before, not employing the established diacrits—their intelligibility to anglophones notwithstanding—for Polish and Hungarian etc results in misspelling. We use the "gli" in Italian, the "ll" in Spanish, all of which are official sanction but foreign to anglophones. Thus, cries that Kunrei-siki is horrendous for anglophones amount to special pleading. To Rhialto, beyond that, I'm far less keen on Kunrei-siki than on Passport Hepburn, which is used by the Foreign Ministry for foreign—especially anglophone—consumption. Why is what's good enough for the Foreign Ministry of Japan not good enough for Wikipedia? My interest is making articles appealing and intelligible to anglophones: We can still use macrons. people, by treating them like IPA pronunciations and use them once in an article. But if we don't want intelligibility or accessibility but instead want "correct" spelling, then we should be consistent with other languages' and use Kunrei-siki.--Monocrat 16:10, 15 November 2006 (UTC)
fwiw, I just finished asking around 3 different schools I work at, and the consensus appears to be Hepburn is the best system to use for consonants, but to write out vowels word processor style.
Personally, regarding consonants, Hepburn is the only sensible option - Huzi is just incomprehensible. What Monbusho (education ministry) says is not relevant, as both their audience and their purpose is at odds with ours. Mofa is more relevant. They require Passport Hepburn (optionally including oh for おう). Hepburn is also the official standard for road signs and de facto standard in many other contexts. I don't know of any source which specifies which, if any, of the Hepburn standards is to be used in official documents.
For vowels, the situation is less clear-cut.dropping the macrons (and circumflexes) was a unanimous opinion in my albeit unscientific survey. The inconvenience of typing these is to me just an extra reason to drop them, and Hepburn does have an option for allowing them to be dropped (vowel doubling). In addition, many schools and offices in Japan, even today, do not have the necessary software installed to display macrons properly. The specific method my co-workers favour (kana spelling for vowels) is in fact recognised as a common variation for Hepburn. I think, if we were voting, Hepburn consonants with kana spelling vowels would be my choice. Rhialto 07:02, 17 November 2006 (UTC)
Nothing personal, but I'm not interested in what is or isn't the consensus among your colleagues, my colleagues, or anybody else's colleagues, at least as long as they don't provide their arguments. What Monbusho (education ministry) says is not relevant: I agree with you there, if only because Monbushō (or however you care to spell it) hasn't existed for five years. (And I thought that I was out of date.....) The inconvenience of typing [letters with macrons] is to me just an extra reason to drop them. If you haven't already set up editor to do this in a more convenient way, copy the little string āĀēĒīĪōŌūŪ into any page you're working on, copy from that to taste. In addition, many schools and offices in Japan, even today, do not have the necessary software installed to display macrons properly. I find this extremely hard to believe. The huge majority of Japanese schools and offices are in thrall to (i) Micro$oft, and (ii) the notion that any old computer hardware is in deadly danger of exploding or being taken over by "viruses". 'Doze 2000 (and NT?) and above handle unicode, browsers for them handle UTF-8, OSes come with fonts that handle these characters. If they're really stuck with some atrocity such as Windows Me, you might suggest Kubuntu: it handles these fonts, and the price is right. -- Hoary 07:25, 17 November 2006 (UTC)
Well, I didnt point out that it was an unscientific survey of my colleagues. I didn't provide their arguments to this forum, as I didn't think they were sufficiently notable to cite formally. But seeing as my colleagues are teachers, it seemed a reasonable start on a direct rebuttal of the claim that only kunrei-shiki is used to teach kids here. As for monbusho, of course I meant the current successor of that body. To insist I meant otherwise is to deliberately find straw men to attack instead of my real point. I hope that wasn't your intent.
The bit about schools and offices is quite true however. Between the schools I work at this year and the 20 or so I did last year, the overwhelming majority still use windows 98, and do not have any font installed that will correctly render a macronned vowel. Yeah, that's just personal observation, but I'd like to see you provide scientific evidence to the contrary if you really are going to insist Japan is fully modernised on this specific aspect. And you're right I could suggest various software upgrades. I doubt they'd listen though. My experience with Japan and technology is that it's a mix of high tech and low comedy.
Between several universities, companies and many individuals, I have not seen a single computer in the last six years using Windows 98 in Japan. As a programmer, I do a lot with computers. Windows 2000 was incredibly popular and since then most systems have moved to XP. I realize that does not mean it does not exist though. However, I must point out that Windows 98 (as well as ME) are dead. Microsoft support, including extended support, is officially over: [2]. Just for fun though I installed Windows 98 SE (English). It installs with IE5. I visited several webpages with macrons and other diacritics. The default install had no problems displaying them. It did have trouble displaying Japanese though. That was fixed by downloading the global IME: [3]. Of course IE6 is available for both Windows 98 and ME, too. Bendono 20:22, 17 November 2006 (UTC)
Official support from product X and installed user base of product X are not the same thing. Maybe it's an issue that appears only for the Japanese version of windows 98? Istr the English version comes with extra support files for most latin-script languages (I know of 2 exceptions). I do know the 2 versions come supplied with different fonts. This macron support is a genuine issue for me; those words with macrons are essentially impossible to read from the computers at the schools I work at. I guess this amounts to a unicode vs ascii debate. There is of course a massive installed unicode user base nowadays, but on the other hand, using macrons is using a very strict version of a standard, which has many more user-friendly standards, mostly for a style issue. It's style vs. accessability. It's worth noting here that most pages that include unicode characters also include a template notice that such characters are include and the page may not display properly.
As more purely observational evidence, all street signs in my area, where romanized, use Hepburn except that vowel length is completely unmarked with either macrons, circumflexes, or doubled vowels. Rhialto 08:36, 17 November 2006 (UTC)
Including long i (ie "ii")? Long i seems to be consistantly written "ii" for both people and place names in and out of Japan. It is only long u (ū) and long o (ō) that is debatable. Bendono 20:34, 17 November 2006 (UTC)
There weren't any examples locally that could test the long 'i' issue. However, Iidabashi is reasonably well known, and the 'ii' spelling is not in doubt there. —The preceding unsigned comment was added by Rhialto (talkcontribs) .
If I understand you correctly, you do recognize long vowels (ii in Iidabashi); it is only macrons that you dislike. Would you be satisfied if (not proposing) ū and ō were written "uu" and "oo"? Does it not seem strange or inconsistent to recogonize "ii" but not other long vowels? Bendono 01:34, 18 November 2006 (UTC)

(breaking indent) As I understand it, the ministry of transport signs show long vowels as "ii" for that vowel (doubling "i"), but show neither macron nor any form of doubled vowel for "o" or "u" (I hae no specific information for a or e). That isn't something I have personally advocated; it's what is used in practice by the Japanese government, or one branch of it anyway. Personally, I'd like to see all vowels written out as doubled as the standard transliteration system, with appropriate exceptions for those words that have an established English spelling already (major cities, official translations of company names, self-chosen names of famous people and so on). The macrons fall foul of teh accessibility issue to me.

It's nice that the latest browsers can handle unicode and macrons specifically, but there genuinely is an established user base in Japan which can't handle macrons, and that user base is of specific relevance to articles on Japan, which would be affected by major use of macrons. Rhialto 05:09, 18 November 2006 (UTC)

You don't need the latest browsers. IE5 was released in 1999. I can test the Japanese version of Win98 next week. I suspect there is something wrong with your setup. There are many free alternatives, both browers and OSs. Win95, 98, and ME are unsupported. Use at your own risk. In any case, how far back do you suggest we go? I used to read websites over telnet. Of course it could not display images or fancy formating. Have you used the text version of Links_(web_browser)? Should we remove all images from Wikipedia? Some people may still not have fonts for Japanese. Perhaps we should remove the {{nihongo}} template as well. I just created a tool to fetch raw data from URLs. While it has Unicode support and displays text just fine, unfortuately it can not render HTML. Should I expect Wikipedia to accommedate me? I couldn't access the Internet from Windows 3.0 (or was it 2.0?). For the sake of those unfortuate souls, perhaps we should take Wikipedia offline and take it to printing presses. Bendono 06:04, 18 November 2006 (UTC)
That is a strawman argument and you know it. Images have methods for degrading gracefully (alt tags to give textual descriptions of the image) when the page is read by older browsers, so even though older browsers can't read images, they do not render the information useless. The system used for macrons, as currently being inserted into pages, does NOT degrade gracefully in older browsers, so the issue is quite different. If some method were provided so they would degrade gracefully without any special effort on the part of the reader, my main gripe against macrons would go away. Rhialto 10:15, 18 November 2006 (UTC)
Well, stepping aside from the issue of computer support: if we are not interested in what our colleagues use, then in what are we interested in? Exact spelling of Japanese in a roman script, as determined by the competent authority? Then per Polish and other Easter European practice, we should employ Kunrei-siki. Something intelligible to all anglophones (natives and learners, specialists and generalist) and which is used by the Japanese government itself for foreign consumption? Then Passport Hepburn. Again, with the possible exception of rail-related articles, I don't see that macrons have any reasonable basis for universal use in Wikipedia's Japan articles.--Monocrat 00:51, 18 November 2006 (UTC)
Personally, I prefer and use Kunrei-siki. It is closer to the phonological (as opposed to phonetic) structure of Japanese. However, that is not the reality that we live in. Either with or without macrons, Hepburn is the de facto standard in published texts. I am interested in spelling Japanese as spelled in professional published texts.
Japanese is a language. It is independent of the Japanese government. Technically, Japan does not even have an offical languge. Likewise, English is language. It is independent of the many, many nations where it is spoken and written. As someone pointed out before, Wikipedia does not regulate between British and American spelling. Spelling is not for the Japanese government to dictate. Bendono 01:25, 18 November 2006 (UTC)
Bendono, I personally agree that languages exist independent of their regulators. But go to France and use irregular forms en masse. People will understand you, as you pointed out earlier, but if you defy the Academy, you risk making yourself a fool. Regulators matter, not least since governments use regulated forms. If there were a popular convention among the Japanese in favor of macrons, I would gladly defer to them, since I feel convention matters more than regulation. While there might not be a law or cabinet order designating it as such, the very act of regulating a language establishes its defacto officiality (slight contradiction there, but...). Who needs to designate a tongue as official when all but a sliver of the population speak it? Now, to your other points. English is entirely unregulated, so there is only popular convention to defer to, and there is no universal convention among anglophones, so Wikipedia doesn't choose between them. Anyway, I can live with Kunrei-siki for the reasons you listed, but I would prefer Passport Hepburn: it's official and the defacto standard in generalist anglophone publications, and if Wikipedia is to counter systemic biases, that should be our criterion. Ideally in my estimation, an article would begin as such: "Hokkaido (北海道 Hokkaidō?) is the northernmost of the four main islands of Japan. Originally named Etorofu, Hokkaido was colonized late in the nineteenth century amid maneuvering between the Japanese and Russian governments..." We would also be following practice at the Arabic style manual, which requires a strict transliteration in the article's lead, and standard translations elsewhere.--Monocrat 15:50, 18 November 2006 (UTC)
Whether it is Hepburn or Kunrei-siki, both systems specifically distinguish between short and long vowels. Whether it be a circumflex, a macron, or two vowels, the distinction needs to be made consistently. As Hepburn is the de facto standard in English texts, then macrons it is. That is the way it is and has been for many years in published English books. If there is ever a general consensus to change to another romanizaiton system, then we'll have something to discuss. By the way, would you please clarify your statement that Passport Hepburn is official? Bendono 16:43, 18 November 2006 (UTC)
Bendono, there is no case for macrons 1) outside of the lead of articles generally, and 2) outside of rail-related articles. None. Yes, Hepburn is a loose defacto in English usage, and academics use macrons, but Wikipedia is not an academic publication. Neither The Economist, the Times of London nor the New York Times uses macrons. Show me an English-language newspaper or encyclopedia or non-academic publication that uses macrons as the universal default for Japan-related articles, and I can point to several that do not. Bendono, are we addressing academics or the general population? Should we tailor this for PhDs or high school graduates?
One more time: Passport Hepburn is official as it is used on official documents by the Foreign Ministry as a standard that is specifically meant for foreign, particularly anglophone, consumption. I find it interesting that there is nary a macron in sight at mofa.go.jp, not even for Messers Koizumi and Aso. (A Google-test turned up exactly zero instances of "ō" on that site.) Western academic practice, i.e. macrons, is simply and utterly irrelevant in the face of 1) the generalist audience of Wikipedia, 2) overwhelming conventional usage by anglophones, and 3) the practice of the Japanese state. The Government of Japan itself doesn't care to distinguish long from short vowels (except EI for えい and optionally OH for おう, and perhaps II for いい, I'm unsure) in addressing anglophones, so why should we?--Monocrat 18:04, 18 November 2006 (UTC)
For god's sake, why is it so hard to understand that this is about accurate spelling? There is no reason for us to stoop to the level of mispelling things to appeal to the general masses; we're an encyclopedia - a scholarly collection of knowledge - and we ought to start acting like one. Just as there is a difference between Grolier or World Book and Brittanica, so there is a difference between random sloppy websites and Wikipedia. Take a look at any scholarly journal focusing on Japan, any Japanese Encyclopedia (e.g. the one published by Kodansha), any work that focuses more on getting things right than on making it accessible to the lowest common denominator, and you'll see macrons or some other appropriate method of romanization.

Arbitrary division 3

I don't care what the NY Times uses, I don't care what 5000 random websites on Google use, and I honestly don't even care what the Japanese government uses. What you propose is deliberate mass mispelling, distorting and perverting words to make them look nicer to Anglophone eyes; you're proposing the elimination of valid and valuable linguistic information. Ningyo and Ningyō are different words, and Ono no Azumabito and Ōno no Azumabito are different people. Spellings like Choshu (ちょしゅ) or Hokkaido (ほっかいど)(instead of Chōshū 長州 and Hokkaidō 北海道) are just as wrong as the Western perversion Iappon, Tokio, Meaco... and yet you're alright with the former, but not the latter, which are seen as old-fashioned, colonial, and deprecated. Why? Why be alright with one form of destructive linguistic bastardization and not another? Again, I point out that any respectable scholarly source will reflect the kana spellings accurately, in one way or another. LordAmeth 20:35, 18 November 2006 (UTC)

We can spell accurately without macrons. This is not a misspelling. The people who understand macrons are knowledgeable enough to look after themselves. The macron form of a subject's name can be preserved in parenthesis as is done with Tōkyō in the Tokyo article. Common usage is a guideline on Wikipedia. A Google test is the primary method given for determining this. The guideline states, "When choosing a name for a page ask yourself: What word would the average user of the Wikipedia put into the search engine?" Jecowa 02:35, 19 November 2006 (UTC)
Again, we disagree. It is inaccurate without the macrons. The {{nihongo}} template is exactly that: nihongo, not English. Also, that template only occurs once a page. Should I have to click on each and ever link just to figure out the correct spelling of other terms used on that page? Google is a great search engine. However, most published books are not available online. Go to a library and read some more books. A great many really do use macrons. It is not something that I am trying to force only on Wikipedia. And as an average user, I actually do search first with macrons. In the past I had very little success; that has slowly been improving lately. In any case, we do have and recommend appropriate redirects. Bendono 02:48, 19 November 2006 (UTC)
The nihongo template seems to contain a lot of English, such as the literal translation and the historical name of Tokyo. Perhaps the nihongo template needs to be modified so non-Japanese words can be included in the same parenthesis without being inside the nihongo template tag. People who don't know macrons also have to click the wiki link of an article to see the pronunciation. Why should the style used on Wikipedia ignore the common usage guideline just to provide a small convenience to the few who understand macrons? I believe "average" refers to the mode in this case. In any case, most average users will not perform searches with queries containing macrons. Most average users would not even know how to type a macron. Jecowa 03:15, 19 November 2006 (UTC)
The nihongo template does not help for terms that are not the main topic of the page. For those words, you would need to click on each and every link to find the nihongo template. A person who does not know macrons (or other diacritics for that matter) does not need to click on extra links to be able to read and understand words with macrons. When I encounter English words like café, crème brûlée, déjà vu, smörgåsbord, or voilà, I do not really know what the diacritics mean. However, I simply ignore them and continue to understand the word. I can not infer those diacritics from their absence. They must mean something to someone. My inability to understand them does not mean that I should try to remove them. Please do not try to speak for others. I can only speak for myself. However, Wikipedia does try to accommadate those who search without macrons with redirects. Bendono 03:32, 19 November 2006 (UTC)
Cafe can be correctly spelled without diacritics. Also café is easier to type than Cafē. The acute accent can be typed with the default U.S. Roman keyboard layout. This is also true of the circumflex, the grave accent, and the umlaut. Typing a macron requires switching to the U.S. Extended Unicode keyboard layout. Don't you believe that most users use their default keyboard layout? Even if everyone did know how to type macrons, most would not know where to place the macrons. As you mentioned above, even you, Bendono, don't know where all the macrons go. The naming convention states, "Since 'Jimmy Carter' is the most common form of the name, it will be searched on more often, and having that exact string in our page title will often mean our page shows up higher in other search engines." The un-macronned form is a correct way to transliterate people's names. The convention says that we want to maximize the number of people brought to Wikipedia by choosing the form they are most likely to search for. Jecowa 04:25, 19 November 2006 (UTC)
I can not (or at least do not know how to) type acute accents, circumflex, grave accent, or umlaut with my default US keyboard. However, you do not need a special keyboard layout: Wikipedia provides them all (and many more) for simple inputting under all edit screens. The above words were taken from English words with diacritics. I did not choose them. I said that I do not know where the acute accent, circumflex, grave accent, or umlaut goes. I do know where macrons belong though. Again, you are trying to speak for others about what is most common. I read through my Kodansha encylopedia and note that without exception all personal names and places use macrons where appropriate. I want to search for more inforation, so I check the same spelling in Wikipedia. Without the macrons, I may not find it. And even if I do, it does not match my sources. There needs to be consistency. Bendono 04:43, 19 November 2006 (UTC)
The New York Times and many other publications have a non-diacritic policy. Kodansha's English encyclopedia consistently uses macrons without exception for places, people, and even so-called "English words". Encarta is generally consistent about using macrons. Here are but a few links:
Academic or not, Wikipedia needs to be professional and spell words correctly and consistently. I gave this sentence earlier for thought: "To day eye went two too stores and bought to packages of meet witch I will cook to knight." Understandable? Yes, with a little extra work. Do I want to read it? No. I do note that all of the words are in the dictionary. Our target should be inclusive of both the general population and academia. It is easy to ignore macrons. However, it is much more difficult to infer a missing macron.Bendono 01:00, 19 November 2006 (UTC)
I'm sorry about saying that you do not know where macrons belong, Bendono. I misunderstood and thought you meant that you couldn't infer macrons and the other diacritics. About the user reading the "Kodansha encyclopedia." Why would this user have trouble finding it without the macrons? Wouldn't the redirects point this user to the appropriate place? How is Wikipedia supposed to match everyone's sources? Jecowa 20:53, 20 November 2006 (UTC)
As we have been discussing, there are multiple forms that these words appear in: with and without macrons. The form appearing with macrons is more accurate. The form removing the macrons is less accurate. There is no resistance to long "ii" ([Niigata], Iidabashi). While it is unfortunate that macrons were decided upon for other vowels, that is the reality that we live in now. If you insist on not distinguishing between long vowels, then should we change Niigata to *Nigata? Or Iidabashi to Idabashi? I want consistency. When there are multiple forms of a word, then I want to use the most accurate form. Less accurate forms should be redirected to more accurate forms, not the other way around.
English words of Japanese origin are not the only words with macrons. A quick search came up with the following subset: Māori, Māori language, Ngā Mānawa, Tāne, Kākāriki, Kākā, Pāṇini, Devanāgarī, Brāhmī, Pāli, Apadāna. These words come from languages that require distinction between short and long vowels in the same way that Japanese does. I am sure that you can find alternative spellings as well. However, it is a matter of accuracy. Wikipedia as a reference. Accuracy and consistency are important. Bendono 00:21, 21 November 2006 (UTC)
You've previously said that you do not support macrons due to phonetic value, so I don't understand how you can employ the argument of distinguishing between long and short vowels here. Also, you successfully chose an article title (the one that starts with a p) that appears to contain a box on my IE on (Japanese) Windows XP. I think that shows exactly the kind of thing that we want to avoid. And separately, the macrons do not render correctly on my up-to-date Mac Firefox either. Dekimasu 04:13, 21 November 2006 (UTC)

(breaking indent) If you want more accurate to Japanese spelling, you should be dropping Hepburn entirely and going for some kind of waapuro romaji. If you want accuracy to official standards actually in use, Hepburn with no macrons, as per ministery of transport, is what you should be advocating, or perhaps Hepburn as used in passports, with that funky 'oh' for おう. If you want accuracy to Japanese pronunciation, you should be advocating IPA, not any kind of romanization scheme. English orthography's most notorious characteristic is an utter failure to bear more than the most superficial resemblance to pronunciation, and trying to force it to is silly and lame. Rhialto 01:36, 21 November 2006 (UTC)

I use what is used in published books. It is not my place to decide to use waapuro rōmaji. Hepburn uses macrons; it does not remove them. Passports do not use macrons because many decades ago some countries could not handle them. Wikipedia is not a passport. I am all for spelling English in IPA. However, again that is not my call. Macrons are not about pronunciation. English spelling is not about pronunciation. Notice that we distinguish between "to, too, two", "night, knight", "which, witch", "be, bee", "wood, would", and many other such words. There is no need for it at present, and yet it is accurate and correct. It is not our place to change that now. The reality is that Japanese terms in English are written in Hepburn. It is fairly consistent, especially when compared to English spelling. Removing macrons introduces inconsistency and is less accurate. It is not our place to suggest spelling reforms. We must work within the present system. Bendono 02:24, 21 November 2006 (UTC)
It's all very well noting that various academic journals and encyclopediae use macrons. But equally, a fair number of atlases do not, and the vast majority of ordinary people, and all of the Japanese government publications and signs, do not. Outside of the ivory towers, no one really uses macrons for Japanese. You're only part-right in saying that Japanese terms are written in Hepburn. Specifically, Japanese, as used in all official literature, uses Hepburn without macrons (or apostrophes); a variant of standard Hepburn. You say removing macrons introduces inconsitency. I say it is inconsistent (and therefore less accurate and introduces confusion) with existing Japanese practice when they write Japanese words in this script of ours. Rhialto 04:33, 21 November 2006 (UTC)
Wikipedia is an encyclopedia. It is a reference. It is important to be accurate and precise. English translations of Japanese literature often do use macrons (Genji and Heike come to mind immediately). Of course it depends on what kind of literature you read, too. The scope of Wikipedia is Wikipedia itself. Whatever we decide here will not change Japanese street signs. Bendono 05:52, 21 November 2006 (UTC)
eta: Even if the vast majority of Anglophone-produced literature used macrons (not true; it's mostly used by academia), wikipedia policy states that self-identification trumps everything else in what name to use. And Japan's current self-identification for all place names is to use Hepburn while dropping macrons and apostrophes. [4] [5] [6] [7] Every government site I can find seems to avoid macrons. That's pretty conclusive as far as self-identification goes. Rhialto 04:54, 21 November 2006 (UTC)
And yet when you arrive at train stations at most of those places you will find macrons. One form is more accurate than the other. We have had this conversation many times now. I do not think there is anything new to debate here that has not already been said. Bendono 05:52, 21 November 2006 (UTC)
Train stations may refer to themselves with the macrons, but as Rhialto mentioned, most (if not all) Japanese government sites don't use them [8]. Thenaming conflict guide says, "Where self-identifying names are in use, they should be used within articles." Also, as Nihonjoe mentioned, The Manual of Style (Japan-related articles) states in the first sentence of the first section, "The en:Wikipedia is an English language encyclopedia. An English loan word or place name with a Japanese origin should be used in its most commonly used English form in the body of an article, even if it is pronounced or spelled differently from the properly romanized Japanese." Where is the guideline that says that words unknown by most English speakers must be romanized using macrons for use in articles? The naming conventions says to "Use the most common name of a person or thing that does not conflict with the names of other people or things." Most Google tests will show that macroned forms are less common than the un-macroned forms. It doesn't matter if editors prefer macrons or what scholarly texts use. Guidelines say to use the most common form. Jecowa 06:45, 21 November 2006 (UTC)
Outside of the exceptions listed in the WP:MOS-JA, the MOS-JA guideline is to be used for all Japanese on Wikipedia. ···日本穣? · Talk to Nihonjoe 07:35, 21 November 2006 (UTC)
日本穣, does this mean that Wikipedia:Naming_conventions and Wikipedia:Naming_conflict may not be applied to the naming of the article titles of any subject whose name originated in Japan? Jecowa 08:09, 21 November 2006 (UTC)
Did I write that? I don't see where I wrote that. Please re-read the WP:MOS-JA. It's very clear on this very subject. ···日本穣? · Talk to Nihonjoe 08:29, 21 November 2006 (UTC)
I was just wondering if you meant that the WP:MOS-JA should be used exclusively or not. I thought your point was that I shouldn't be trying to apply those guides. I don't understand your point. Could you please go into more detail on your 08:09, 21 November 2006 (UTC) post listed three comments above this one? P.S. Thank you for arbitrary divisions. Jecowa 08:40, 21 November 2006 (UTC)
No guidelines on WP are used exclusively of other guidelines. As I said, outside of the exceptions clearly spelled out on the the page, WP:MOS-JA covers all Japanese usage on Wikipedia. That's the whole point of the guideline: to provide clarification on Japanese usage in articles. In this case, these guidelines are very clear about when macrons should be used, and when it's acceptable to not use them. I really don't understand why every few months someone goes ballistic over something when it's already covered by the MOS-JA. Please read it carefully. This discussion is already covered there. ···日本穣? · Talk to Nihonjoe 09:20, 21 November 2006 (UTC)
As a further clarification, the original complaint in this insanely long thread was that "the Japanese people themselves rarely use macrons." This is irrelevant. This WP is the English Wikipedia, and the style guidelines specified by MOS-JA only apply here, not on the Japanese Wikipedia. Therefore, whether the Japanese use them or not is irrelevent to most or all of this discussion. These guidelines are meant to help accurately reflect Japanese using a modified Hepburn romanization scheme. Those who understand the meaning of the macrons will find what they need, and those who don't will either click on the little question mark or ignore the macrons altogether.
This is a win-win situation. Sure, it may be a little (only a very little) more difficult to enter a macronned letter, but the end result is an encyclopedia that is useful to both the lay person and the person with a more in-depth understanding of Japan and Japanese. Wikipedia aims to be useful to as wide an audience as possible, and this accomplishes that purpose. If you read through the archives, you'll see that most (if not all) of the arguments being brought up here have been discussed to death multiple times. The current state of the MOS-JA was arrived at after intense discussion and debate, and it was determined that:
  1. There is no logical reason to avoid using macrons as the wiki software now supports them (for over a year now).
  2. Almost all browsers and operating systems in use for the last several years support the use of UTF-8 characters, and can display them with little or no adjustments by the end user.
  3. Using the macrons provides additional information for those interested, and the macrons can be (and will be) ignored by anyone else.
  4. Using macrons is not difficult as all of those used for Japanese transliteration are available underneath every edit box on the site.
Even with all that, exceptions have been built in to the MOS-JA to accomdate words which are commonly used in English without macrons. It's actually fairly well thought out. On top of that, all this bickering over personal preferences (some like macrons, others don't, and others are indifferent) doesn't help the project any. A lot of time was put into the current MOS-JA, and so far, I haven't seen a solid, valid reason for changing the part which addresses macrons. Discussing whether individual exceptions should be made is fine, but attacking the hard work of a large number of editors who spent months (and some of them, years) working things out to cover the largest number of possiblities without making the MOS-JA so huge and unweildy as to render it useless—that's just completely unnecessary.
What is all boils down to is this: There are exceptions to the MOS-JA, and there will likely be a few more added as time goes by. However, scrapping the whole of the macron usage part of it is completely pointless and counterproductive. Yes (as I indicated before), entering macrons takes a very slight amount of extra time, but in the end, it produces an encyclopedia that is useful to a larger number of people. Again, that's the whole point of Wikipedia: to be as useful as possible to the largest number of people as possible. ···日本穣? · Talk to Nihonjoe 09:42, 21 November 2006 (UTC)


I like your numbering system for different points, 日本穣. As you can see, the subject of a pair of the same numbers is the same. For example, "2" is about macrons in computer compatibility, "3" is the usefulness of macrons, "4" is about the difficulty of typing macrons, etc.
  1. There is no good reason to include macrons in article titles. The best I've heard is to allow those that understand to view the pronunciation without clicking the link to the article.
  2. Not all computers display macrons. It looks terrible to have the title of an article garbled for the users without computers with that ability.
  3. Macrons give extra information those who understand them. This information can be preserved in the first sentence of an article as is done in the Tokyo article. This is good enough.
  4. Entering macrons can be difficult. The characters in the character input table are too small to allow easy distinction between characters such õ and ō, which look nearly identical (if not identical) at this point size and font. The point size used for the text-input box is slightly smaller than the size in the article.
  5. MOS (Japan-related articles) says to use the "most commonly used English form." This is not usually the macron form.
  6. Wikipedia:Naming conventions (common names) says, "When choosing a name for a page ask yourself: What word would the average user of the Wikipedia put into the search engine?" Without the character input feature of Wikipedia, many users would not be able to search for with macrons.
  7. Self-reference, as Rhialto mentioned - most cities refer to themselves without the use of macrons in their English translations. I also believe that most Japanese people write their own names using English without macrons, but I don't really live around a lot of Japanese people.
Wikipedia article titles are supposed to reflect common usage, not prescribe more accurate names. Macrons lower the usefulness of Wikipedia by lowering the search rankings of the non-macron titles (which would only be redirects to the macron titles). Most people do not type macrons. This is why the default keyboard layouts don't allow for the typing of macrons, because most people don't use them. The majority of people will be searching in their search engines and the Wikipedia search box without macrons. Jecowa 10:47, 21 November 2006 (UTC)
Thank you for the response, 日本穣. It was very well worded. As anyone reading my responses should know, I am very much in favor of macrons. I have continued to counter such comments as above because I felt that I needed to defend macrons. I know I have responded to each of those points already and there really is not anything new to say. As you wrote, as long as there are no serious thoughts being given to changing the guidelines, then I willing to give it a rest for a bit. If in the near future serious consideration is given for changing the guidelines, I would appreciate a notice. Thank you. Bendono 11:31, 21 November 2006 (UTC)
Around and around we go (for as long as needed). Self-identifying names can be taken from the stations as well. Ōsaka / Osaka or Tōkyō / Tokyo is not a foreign loan word; it is a transliteration. You could translate it to "Great Hill" or "Eastern Capital" for English. A reasonably good example of a Japanese foreign loan word is "soy", which is to be preferred to the Japanese term shōyu. You are abusing the style guide. Bendono 06:57, 21 November 2006 (UTC)
You know, the obvious conclusion to draw from that supposed contradiction is that the stations identify themselves with the macroned forms, and the city, as a distinct entity in its own right, self-identifies using the non-macronned forms. they are two distinct entities, and just because one identifies itself in one way doesn't have any bearing on how the other identifies itself. Rhialto 07:40, 21 November 2006 (UTC)

Time for an RfC?

I absolutely concur that there is no reason why macrons or circumflexes cannot be used in Wikipedia. I simply feel there is no logical reason for every instance of a word has to have diacrits when neither standard English usage nor official Japanese policy (again, defer to the French Academy, defer to the Japanese Cabinet) mandates it in this context. I think it is especially unproductive to redirect time and energy to making what I view as superfluous, hypercorrective changes to numerous articles, especially after the policy necessitating those changes has come under concerted attack. I have never suggested abolishing the macron in Japan articles. What I have suggested is that we follow practice at numerous fine anglophone publications and at the online English editions of the Asahi, the Yomiuri and the Foreign Ministry. This would generate policies parallel to those at MOS-Arabic and MOS-Chinese, in which the ultra-correct forms (academic transliterations and Hanji, respectively) are used only once or as needed in an article. A complaint about clicking links to find diacrit forms? Well, 1) there is a Find function on your browser, 2) per my suggestion, my diacrits would be in the lead, so you could just hit the Home button, 3) and tabbed browsing permits ready access to linked topics.

In any case, no one is suggesting that Wikipedia be anything other than professional, and frankly I dislike the imputation that I advocate sloppiness. The NYT is not sloppy. The Economist is not sloppy. The AP is not sloppy. The Foreign Ministry of Japan is not sloppy. Passport Hepburn is in no way, shape or form a misspelling, and the fact that it looses some phonetic information does not mitigate this truth. To borrow from Bendono: "Knight" reflects the sound of its word rather less well than "Taro Aso" does the name of the minister. ("English kehnigits," anyone?) Yet why is the first correct while the second, supposedly, is not?

I too want professionalism. We simply disagree on Wikipedia's profession: the academic knows enough to take care of himself (to borrow the blithe wording from MOS-JA; why make articles any more foreign to generalist readers? With all due respect to Nihonjoe and Bendono, I admire your tenacity here and your contributions elsewhere, but I think it would be appropriate to submit a Request for Comment to obtain outside opinions. I will be on Wikibreak over the coming days, but I will pursue this.--Monocrat 14:26, 21 November 2006 (UTC)

I also don't think the NYT, Economist, or AP are sloppy, but rather they are uninformed or unaware, or that their specific manual of style discourages the practice of macrons. I think the latter is the most likely as it makes it easier for them to write about things if they don't have to bother using macrons or other diacritics. Chances are, they also don't use most of the diacritics used by languages such as Vietnamese and many of the Eastern European languages, yet we use them here on Wikipedia in many or most cases (depending on the topic and the language in question). The same thing goes for the various English-language papers in Japan: the style manual they use likely doesn't indicate they should use them, so they don't.
The case of Taro Aso falls under the exceptions built into the current WP:MOS-JA as Taro Aso is the most common way to write his name in English due to his prominence. I don't really think an RfC is necessary here. ···日本穣? · Talk to Nihonjoe 03:07, 22 November 2006 (UTC)
"there is no logical reason for every instance of a word has to have diacrits when neither standard English usage nor official Japanese policy ... mandates it in this context." – Yes there is. When reading or quoting text from an article, diacriticless words convey wrong or incomplete information. Also, there is a standard English usage, namely the encyclopedic and scholarly, that aplies here. This is an English encyclopedia, not a newspaper, passport, road sign or Japanese government publication.
"2) per my suggestion, my diacrits would be in the lead" – Only some of them would; the others needed would be in arbitrary places around the article or on some other page. Points 1 and 3 in your argument are totally against accessibility and usability.
"The NYT is not sloppy. The Economist is not sloppy. The AP is not sloppy. The Foreign Ministry of Japan is not sloppy." – Yes they are. They have just adopted a more sloppyness-tolerant approach that works well in their environments, but not in ours.
"Passport Hepburn is in no way, shape or form a misspelling, and the fact that it looses some phonetic information does not mitigate this truth." – Yes it is. It is a required misspelling for the international passport systems to be interoperable.
"why make articles any more foreign to generalist readers?" – We are dealing with a foreign language here. Unambiguity and comprehesibility are more important when the subject is inherently "foreign". Generalist readers can either ignore the macrons, like they probably do with other diacritics, or find out what they mean and increase their understanding. Wipe 15:50, 25 November 2006 (UTC)

A (hopefully noncontroversial) proposal not connected with the "what the hell" debate

I want to rename the article Kami-Ījima Station with the new title Kami-Iijima Station. This would put it in accord with Wikipedia:Manual of Style (Japan-related articles)#Romanisation #1: "All other long vowels are written without macrons: ああ → aa, いい → ii, and ええ → ee." However, when we recently changed Wikipedia:Manual of Style (Japan-related articles)#Romanisation Article titles #2, we simply specified "Article titles should use macrons" (allowing the exceptions for English custom). This could be interpreted to require i-macron instead of ii in titles, making it different from body text. I don't believe the intent of the change to macrons in titles was to change this.

So I propose to change Wikipedia:Manual of Style (Japan-related articles)#Romanisation Article titles #2 to read as follows: "Article titles should use macrons as specified for body text except in cases where the macronless spelling is in common usage in English-speaking countries (e.g., Tokyo, Osaka, Sumo and Shinto, instead of Tōkyō, Ōsaka, Sumō and Shintō). I've written the proposed change in boldface to make it easy to see, but of course I don't intend to make it boldface in the MoS.

I hope the insertion of "as specified for body text" will not be controversial. But if anyone has an opinion, please speak up.

Fg2 20:53, 21 November 2006 (UTC)

I believe when it says to use macrons, it means to use them in the manner described above. If the original title was in kanji, then this article would need to use "Ii." "Ī" should only be used when transliterating katakana and only when an English spelling isn't available. Jecowa 21:48, 21 November 2006 (UTC)
It sounds fine to me, and I hope along with you that this can be solved quietly. I don't know if it's explicitly stated in our policies, but I don't think I have ever seen macrons used in place of ii, ee, and aa. This should be perfectly fine... I don't want to see "ē ja nai ka" or "Īdabashi" or "kakkō ī". That's just really really weird. LordAmeth 23:19, 21 November 2006 (UTC)
Jecowa is correct. The correct title of the station would be Kami-Iijima Station. Only double katakana is written with the macrons in all cases. ···日本穣? · Talk to Nihonjoe 00:21, 22 November 2006 (UTC)

Looks like a redirect may have already been there. I was bold and just moved the article and cleaned up the links to it. Without the macron seems to be the only way it's listed as far as I can tell when doing a search. With the macron yeilds no results. I excluded Wikipedia and all mirrors in these two searches. ···日本穣? · Talk to Nihonjoe 02:50, 22 November 2006 (UTC)

Thanks to all Fg2 07:21, 22 November 2006 (UTC)

A sidenote for the letter "I" with macrons... Whatever works for Polynesian languages don't necessarily work for Japanese. For example Hawaiian language has macrons for "I", as in Waikīkī. But because of several "pronunciation respelling for English" notations, Waikīkī can be misread as Waikaikai. And Ījima can be misread as Aijima. That's very confusing and unnecessary for Japanese. For Japanese related articles, I believe we should avoid "I" with macrons because of this confusion. (However, no need to mention this in the MOS-JP).--Endroit 17:05, 22 November 2006 (UTC)

Not again...

Away for a little while and there it is again. These macron debates are very repetitive and always end the same way. Perhaps at the top of the talk page there should be a link to a "macron FAQ", explaining why we use macros, that yes, your question has been asked before and no, we won't remove them. And apart from all the reasons to use macrons, people should also come to realize that there are a lot of Japan-related articles now, and that changing our romanization style (either to waapuro, non-macronned style, or whatever else) will involve an incredible lot of work. After that is done, there are probably articles we missed, hence still using the style we use now. Net result: no gain whatsoever and loss of consistency. People should be able to put this issue to rest. And I say this as someone who uses a hyphenated waapuro in his private life, so no, I'm not just defending my favourite system. Shinobu 23:57, 6 December 2006 (UTC)

The person who'll write that FAQ deserves a barnstar (or sensu). It won't end the eternal debate but will give it a decent home. Wipe 01:44, 7 December 2006 (UTC)
I started one a little ago. I never made it very far, but if there is interest, then I can try to put at least a rough version up this weekend or next. Bendono 03:08, 7 December 2006 (UTC)

If you put it up, I will read through it and see if I can add anything, referencing the old talk archives where possible. Shinobu 07:22, 11 December 2006 (UTC)

Sorry but....

Sorry but permit me to introduce an Extended Hepburn System, a transliteration system of Kana syllabary into Latin alphabet. It is devoid of any diacritical marks but produces good result when applied to the classical Kana script. Please read the following.

Talk:Romanization of Japanese#An Extended-Hepburn System
ja:ノート:ローマ字論#擴張ヘボン式

Kmns tsw 08:50, 17 December 2006 (UTC)