Wikipedia talk:Manual of Style/Japan-related articles/Archive 6

using hyphens in name

hello everyone, I don't know if there is a rule for this or not, but I have a question regarding the hyphens in names "xxxx no xxxx". for example it's Ninigi-no-Mikoto with the hyphen, but it's Soga no Kitashihime. so what is it? I personally prefer the version without hypens, since that would come more closer to the original japanese version, but I was wondering what others think? also one more question, so is it settled now with the macrons, they will be used on all japanese words and terms? except for example for Tokyo or Shinto, etc...? domou arigato. Gryffindor 14:27, 2 March 2006 (UTC)

I would personally prefer no hyphens. This hasn't been set in stone, however. I'd say that unless it's a combination last name (e.g., "Smith-Jones"), no hyphens should be used. As for the macron issue, we're just waiting for the mediation to take place. There's a large backlog on the mediation page, so it may be a while. If you go by the discussion above, however, the majority of people seem to accept the use of macrons, and the MOS-JA has been changed to reflect that. --日本穣 20:33, 2 March 2006 (UTC)

My thought on it is that if it's a divinity (Ninigi-no-Mikoto), place (Clacton-on-sea) or hyphenated name (Smith-Jones), then it goes hypen. Otherwise, don't bother with them. As for macrons, I personally prefer them, but it's currently awaiting Mediation.--み使い Mitsukai 14:47, 3 March 2006 (UTC)

Recent discussions were about macrons in article titles, not macrons within the text of articles. Macrons within articles were fully established in or before 2004, but macrons in article titles were not then possible and that's the issue that we discussed a month or two ago. Within articles, long-standing policy is to use macrons for words and names that aren't fully established in English. As Gryffindor notes, "Tokyo" and "Shinto" are examples of words that we write without macrons. Fg2 22:00, 2 March 2006 (UTC)

Right, that's what I meant, if macrons are used in the article titles. I would support using them, is there a place to vote or has that been closed already? And also the use of hyphens (or not using them basically) I guess we can discuss that here, right? Gryffindor 22:44, 2 March 2006 (UTC)

Yes, it was discussed multiple times, and a decision was finally made and is reflected in the current WP:MOS-JA. Hopefully that will answer your questions regarding macrons in titles. (^_^) --日本穣 22:59, 2 March 2006 (UTC)

Ah I see so for example articles like junihitoe and Empress Kojun can be moved now to macros? or do they need to be submitted first for a move discussion? Gryffindor 14:06, 3 March 2006 (UTC)

The last I heard, the macrons-in-titles debate was still undecided as we await Mediation on the matter. I'd wait until that process finishes before any wholesale moves get underway. As for the titles of new articles, I'd say do what you feel comfortable with. I prefer macrons, myself. — BrianSmithson 14:20, 3 March 2006 (UTC)

Yes, I think it is good to wait for the mediation before making changes to existing articles. For new articles, if you macron the title, be sure to create redirects from the unmacron'd version as well. Neier 15:58, 3 March 2006 (UTC)

Wakarimashita. ok, so now for the hyphens. yes or no? I say no hyphens, what do you folks think? Gryffindor 18:26, 4 March 2006 (UTC)

@I personally prefer the version without hypens, since that would come more closer to the original japanese version, but I was wondering what others think?: Actually, the "no" is not written in the kanji versions in the articles. Since "no" is a genetive marker (particle), or at least that's what I think it is, it would make sense to spell it without hyphens, because that's what everyone else seems to do. (Even though I prefer to have a hyphen before particles, but I know that my taste in this matter is pretty unique.) Shinobu 02:56, 5 March 2006 (UTC)

Non-breaking Kanji

How can we make strings of kanji non-breaking? That is, how can we prevent the browser from breaking between characters at the end of a line? I've tried using zero-width non-breaking spaces (), but it doesn't appear to work. Dforest 06:38, 12 March 2006 (UTC)

I guess the first question is, why?-Jefu 07:53, 12 March 2006 (UTC)

I was afraid someone would ask that! ;) Basically I just want the browser to treat a word of several kanji the same as an English word, and not break it in the middle arbitarily. This usually isn't a problem to have the word broken in purely Japanese text, but in the English Wikipedia, where we sometimes use foreign scripts within English text, it could make it awkward to read.

For example:

The Japanese word for Japan is Nihon 日本, meaning "origin of the sun".

might be broken by your browser as:

The Japanese word for Japan is Nihon 日

本, meaning "origin of the sun".

--Dforest 08:32, 12 March 2006 (UTC)

Bear with it, I guess. These things happen. --Golbez 09:19, 12 March 2006 (UTC)

Use the nobr template if it is really awkward. For example, Nishida's World-of-worlds formationism:

This is a very long line isn't it? The last word might break! 世界的世界形成主義.

It won't break. :) JeroenHoek 10:45, 12 March 2006 (UTC)

Thanks! --Dforest 11:00, 12 March 2006 (UTC)

Macrons

I see much use of macrons above o characters in Romanized Japanese on Wikipedia. Some people will go to a page and change the ou's or oo's to ō's. While this "works" and may be the "official" way to translate using certain methods, I see this as only contributing to faulty re-translation. One cannot tell whether to write oo or ou when using kana. Even while using the internet and roomaji still, few people know how to make a macron. So, for most people, they will attempt to add another o or a u (likely with faults) if they don't simply leave off any extra markings or characters entirely. We're probably all more familiar with romaji than roomaji, for example. How would you know to write a second o unless you had encountered it in Japanese before? Few people know Toukyou, and Tōkyō doesn't tell one how to write it properly with phonetic Japanese characters. When these writings are reproduced in most places outside of Wikipedia, they will lose their macrons very rapidly. Even if they don't, the next person to use them is likely to remove them also. I think that a person is more likely to leave on a u or second o than the macron above an o. I also think that the majority of people who use computers don't know how to make a macron to save their life. Beyond this, I imagine that some percent of people narrowly bordering 100% have no idea how to make a macron by heart (without access to some list or copy-pasting). If anyone agrees or disagrees with this, let me know. 72.145.133.228 11:14, 18 March 2006 (UTC)

"Wikipedia uses the Hepburn romanisation because it is generally accepted by scholars and it gives a fair indication of Japanese pronunciation to the intended audience of English speakers." (quoted from this MoS)

While I generally use waapuro for personal romanization needs, for an encyclopaedia it is important to use a romanization system that is well established among scholars and gives a good idea of the pronunciation. I think I can safely say that in scholarly circles Hepburn is the most widely used system, and it gives a fair idea of the pronunciation. Hepburn uses macrons, so we use macrons, although when there are common alternative spellings, we mention them (e.g. Shounen).

I agree with you that macrons are at greater risk of being omitted, but on the other hand "oo" and "ou" are more at risk of being pronounced like the "oo" in "book".

Entering macrons in articles is very easy, just click the ō below the edit box. As for a list, every computer has such a list. It's, depending on your OS and language, usually called something like "Special symbols" or "Character map" or something similar. Click it and type "macron" (or the equivalent in your user language) in the search box - a list of symbols subborted by the selected font will appear. But even if it's very easy, that doesn't mean that people can do it (after all, most people have a room temperature IQ). So yes, the macrons will probably disappear.

Then again, Wikipedia is not here to teach people Japanese. Those who're interested will mark the macrons, those who're not will screw up however you spell it. It is also important to note that ō is not rendered "おう" or "おお" most of the time, because in Japanese words it's usually part of a kanji (like 凹 or 大) and in foreign words a lengthening mark (as in ローマ字). Shinobu 17:11, 18 March 2006 (UTC)

I think an important thing to note is that while Japanese people sometimes use macrons, they already known the true Japanese versions of the words; thus it is not as important for them to have an indication of exactly what character is used. At the least, it would be nice if a phonetic Japanese version of the words would be available with words. It could be noted, however, that when ou and oo are used, the kana version may not even be needed at all. And certainly, we could say, "A person will pronounce it wrong if we have an 'ou' in there," but a person could pronounce arigatou as uh-rig-uh-toe, macron or no macron. While Wikipedia is indeed not "teaching" Japanese, it is doing two key things. One, it is often introducing many people to Japanese, and two, it is also providing a reference for what certain words mean and how to say and use these words. It seems rather important that it doesn't assist the spread of misinformation. While arigatou may have it's faults, arigatō does too, and one is not helping stop arigato from being more popular on the Internet. Arigatou is commonly written out with hiragana in Japanese, so there is a problem. With pronunciation, kimono tends to be pronounced differently than the Japanese version (even to the point that dictionary pronunciation keys can show only the wrong way), and the r in arigatou is really not an r. While making use of ou and oo to show which character should be used by a person writing it in kana is not necessary, neither is showing the original Nihongo version. I don't imagine that the lay-person is going to have much of an idea what the line above an o means or what macron means (or to search for "macron" in the character map.. or remember it or to bother to take the time to do so, of course). Even if they happen to be big on proper pronunciation, the r is not properly pronounced as an r.

In the end, we could add markings with the other vowels and use a character besides r, but we don't. Why be so worried that, in this one case, pronunciation will go so wrong? Why take that to the point of intentionally damaging what would otherwise be an incredibly reversable translation system and throw a wrench into it? Why not use something like "ōo" and "ōū"? Why not always include kana in the little group of parenthesis after the word?

p.s. I worry you're a fan of Celsius. ;]

72.145.133.228 20:38, 18 March 2006 (UTC)

Proposing some non-standard way of transcribing a language for an encyclopedia is just silly. Just because a bunch of anime fans feel most comfortable using wāpuro rōmaji doesn't mean Wikipedia should just use that. Almost all of the academic research done in English on the subject of Japan and its culture uses macrons. (Where macrons aren't used the text in question often uses the circumflex (because the text was written with a font that did not support macron vowels), but this is not necessary anymore.)

More importantly, anyone who needs to know the exact spelling of "Arigatō" already knows (or should learn) how to look up such a word in a dictionary! The Hepburn system enables anyone to pronounce the Japanese word, which suffices quite nicely for an encyclopedia. JeroenHoek 21:09, 18 March 2006 (UTC)

That doesn't really address my ultimate point though: Why bother with an layman English version of the word along with a "proper" (according to some) pronunciation key which is only properly pronounced if one already knows how to pronounce Japanese characters written in that form of Romanization? If a person has to know how to say an i in Roomaji, not to mention an r, how helpful is it to have a macron? Unless a person knows how to pronounce Japanese, it's more likely than not that they will pronounce the normal or elongated vowels too long. They also need to have knowledge of the structure of Japanese in order to properly group sylables together.

Basically, I agree with the "Oume-shi" decision (far) above where ou'me'shi would be annoted as:

Ōme (青梅市; Oume-shi)

Also, I think that kana are a good idea. As an example of ideas, rather than:

Junichiro Koizumi (Japanese: 小泉純一郎, Koizumi Jun'ichirō, born January 8, 1942)

there could be something such as one of:

Junichirō Koizumi (Japanese: 小泉純一郎、 こいずみ じゅんいちろう、 Ko'i'zu'mi Jun'i'chi'rō, born January 8, 1942)

Junichirō Koizumi (Japanese: 小泉純一郎、 こいずみ じゅんいちろう、 Ko'i'zu'mi Jun'i'chi'rou, born January 8, 1942)

Junichirō Koizumi (Japanese: 小泉純一郎、 こいずみ じゅんいちろう、 Koizumi Jun'ichirou, born January 8, 1942)

Junichirō Koizumi (Japanese: 小泉純一郎、 こいずみ じゅんいちろう、 born January 8, 1942)

I think these are easily approachable by those who are not familiar with Japanese and very helpful to them. Those who are seeking to learn more about basic Japanese are also further informed (regardless of the depth of their knowledge or interest). Those who are are very into phonetics have even more information (removing the need for them to research just to figure out how to pronounce a single word).

I think that the simple inclusion of kana eases any confusion by those who already understand the phonetics of Japanese, though more could be done to aid pronunciation with Romanization. The system being used is for people who already understand how the sounds of Japanese are organized and how to pronounce them (with simple assistance for the n's). A person is much more likely to 1) not understand the macron or markings, 2) not understand the pronunciation of the letters, or 3) understand how to pronounce the kana already. I, at least, see solutions such as these as solving whatever problems a person might have with getting enough information, and the only ones I can see as being dissatisfied are those who rigidly stick to standards merely for the sake of the standards. One mistake which I believe is often made by major contributors is that they design for themselves. Wikipedia is not just about experts with developed, advanced skills. Wikipedia can be technical, but it is also supposed to be accessable to the general public. Say, "Anyone who needs to know the spelling of arigatō can look it," all that you wish, because the same person could look up just the kanji of it as well. So why give a pronunciation key which is only helpful to the specific group of people who know Japanese Romanization well but don't actually know kana? 72.145.133.228 05:41, 19 March 2006 (UTC)

We've already answered the question: Because that's the accepted scholarly system in use for the English language. If Wikipedia is to be a scholarly source, we need to follow suit, not try to blaze a new trail and reinvent the wheel. And the number of people who know about Japanese topics but who don't know about kana is probably higher than you think. I've been reading books on Japanese mythology, religion, and art for about ten years now, and I've just now started learning written (non-romanized) Japanese. The pronunciation issue isn't a big one for those works, since most of them include a note in the introduction about the Hepburn system and how to pronounce the macronned vowels, the r, the n, etc. This is a problem for any language being read by an English speaker; how many people know how to pronounce the French ç or the difference between the French é and è? How many people read haïr as hair rather than ha-eer? Ignorance of the masses should not be an excuse for Wikipedia to change from accepted academic practices. That said, of your suggestions above, I wouldn't have any problem with the first option (kanji, kana, Hepburn). Others may argue that it's too many foreign characters in the introduction, though. — BrianSmithson 16:03, 19 March 2006 (UTC)

Certainly a person needs to learn how to pronounce letters in a foreign language when they are learning it, but that is really my point. A person can learn "ou" and "ei" just as easily as they can macrons. It may be acceptable by some linguists to use that method, but that is just for pronunciation alone. Hepburn works well for that, and I think that it's fine for the basic entry for laymen to use it. However, it doesn't indicate what sounds go with what kanji, what kanji go with what words, or whether to use oo or ou. Thus, you can't easily use the kanji of a person's given name or surname individually, use kanji individually, learn about the kanji, or properly and easily enter many kanji with a QWERTY keyboard or even a kana keyboard. Hepburn works for pronunciation of a limited number of words far better than it does for truly learning, understanding, or typing the language. 72.145.133.228 22:23, 19 March 2006 (UTC)

Whichever pronunciation key we use, the reader will always have to familiarize himself with it.

Having both kana and romanization would be unnecessary bloat, as they convey pretty much the same information. Most of our readers can't read kana, ergo the use of romanization scheme.

Regarding the Koizumi example, I think the current representation, Junichiro Koizumi (Japanese: 小泉純一郎, Koizumi Jun'ichirō, ...), serves its purpose just fine. It lists the customary English spelling, the Japanese spelling and the pronunciation. Having kana is redundant. Listing another pronunciation key while we already have a good pronunciation key is also redundant. I don't think extra apostrophes will help the layman a bit, and also try to imagine how 小泉純一郎、 こいずみ じゅんいちろう looks to the layman: XXXXX XXXX XXXXXXX.

Currently the pronounciation key given is used in academic papers and books for a reason: it's good. There currently is no nead to research to figure out the pronunciation of a word or phrase: it's fully contained within the romanized form.

As for people not understanding the romanization, I'll think that problem will not only not be fixed by adding apostrophes and changing macrons to o's and u's, but it will not be that dramatic either. Every normal English dictionary has a system of giving the pronunciation too, with the pronunciation key given in e.g. the inlay. People can be assumed to be familiar with these - to them, getting familiar with Hepburn will be easier than getting familiar with kana.

Another nice thing about Hepburn romanization is that it's very accessible to English speakers. Hepburn can be mastered by a layperson litterally within a minute (that's personal experience speaking). We're not using kunrei for a reason.

To sum it up, I don't see how the benefits (if any) of using waapuro, weigh up against dropping Hepburn. Remember Hepburn is used in virtually all scientific papers and books, books on Japanese culture, and Japanse courses, including those targeted at beginners. I also think that giving Japanese spelling, pronunciation and common English spelling, if different, is enough. Anything more will not really be useful and just lead to article bloat.

If the nihongo template gets a bit more accepted and versatile, we might be able to accomplish things like "kana if the user wants to" though. We needn't provide the kana though, because most kanji have either "ou" or "oo" spellings and the "oo" are few enough to script all the exceptions using user script. But that's far in the future, if ever.

Oh, and by the way, you say "the r in arigatou is really not an r". Well, here's something: the r in English is really not an r! To me, the r as commonly pronounced in anime and Japanese songs sounds like an r. Of course that doesn't really help English speakers, but apparently Hepburn felt that it was still better to use the r for this sound.

And yes, I am a fan of Celsius, and proud of it. (I'm fan of Kelvin too, of course.) When (not "if", "when") the entire world uses S.I. for everything we will be rid of one big problem indeed. (note: this post was written before having read Brian's post) Shinobu 16:19, 19 March 2006 (UTC)

(reading Brian's post didn't prompt changes to mine) Shinobu 16:22, 19 March 2006 (UTC)

Page title: Junichiro Koizumi

Jun'ichirō Koizumi (Japanese: 小泉純一郎、 こいずみ じゅんいちろう、 Ko-i'zu'mi Jun-i'chi-rō, born January 8, 1942)

Jun'ichirō Koizumi (Japanese: 小泉純一郎、 こいずみ じゅんいちろう、 Ko.i'zu'mi Jun.i'chi.rō, born January 8, 1942)

Jun'ichirō Koizumi (Japanese: 小泉純一郎、 Ko-i'zu'mi Jun-i'chi-rou, born January 8, 1942)

Jun'ichirō Koizumi (Japanese: 小泉純一郎、 Ko-i'zu'mi Jun-i'chi-rō(u), born January 8, 1942)

I was thinking about it and wondering how acceptable either of these would be. The page title would be easily enterable. The initial name would be simple but very helpful for pronouncing to someone who has basic knowledge of Hepburn or Japanese while not being terribly intimidating. The kanji are clearly unique of the phonetics and helpful, and the kana provide a pure representation of it in Japanese without confusion over ou/oo. The Romanization following those provides a highly helpful version which is broken down into both it's kanji/kana phonetic syllables (with ') and apart at the separations between kanji (with - or .). It would be nice to seperate using hyphens between kanji, hyphens between kana from kanji that aren't connected [not okurigana], and periods between kanji and kana that are connected. This would allow a person to know which part is a surname, which part is a given name, and the pronunciation for each individual kanji and word.

Now, when there is no n before a vowel, the initial writing would only have macrons or neither. When there is nothing that kana can clear up (ou vs. oo), I think omitting them would be perfectly fine. If there is only one kanji with simple phonetics or only two simple syllables and two kana or some other non-ambiguous version, I think that omitting even the highly phonetic version would be fine. Most of them would not have ou/oo or anything such as that, removing any need for kana. So while it would complicate some, it would simplify others.

Do note, I don't mind the page title having a macron, though that would lead to more re-direction. It also provides a simple, alphabetical version of the topic (with no 's or macrons). That's really another discussion though (already mediated above).

And to respond to your message: Yes, most cannot read kana, but isn't it helpful for a person to understand that ou/oo does make a difference in Japanese rather than teaching them a word with piece of it missing? If they don't care, they can always just remember how to say it, but I think giving them access to that information is a good thing. I think it would be easier to just eliminate macrons from the second romanization than to use kana just to clear up ou/oo kinds of issues. Just because Hepburn is easy doesn't mean that it provides all of the information it should. In fact, it being quick implies quite the opposite. While it may take only a moment for a person to learn that, "Those lines above vowels are 'macrons' and they mean the sound is longer," it takes a similarly short time to learn that, "Two vowels in a row mean a longer sound as does ei stretch e and ou stretch o." Now, to those who preach formality in using the Hepburn system, kana (or something else) need to be used in order to express the difference between oo and ou. The first part of this would be Hepburn (or whatever Romanization might be in in place for it), and that should be helpful for advocates of it, and then the second part provides not just phonetics but also information about the kanji, words, Japanese name order, and full pronunciation (further shortening the learning curve of pronouncing Japanese).

I didn't know Wikipedia to be one to shy away from doing things a bit differently if it means getting better information out there. In fact, Wikipedia seems to thrive on reinventing the wheel. If there's resistance to that, the first two keep Hepburn-like pronunciation, though the last two are much shorter. So there is the "benefit" of using waapuro, a shorter way to distinguish ou from oo, not to mention that we are using computers which currently enter it much more easily. Of course, we could always come up with something unusual like,

Tō-kyō (ou, ou) or Tō-kyō (ろう、ろう)

but I personally think waapuro is fine. It is easily learned, is very common, is easily understood, and reduces the death of a second o or u, and plus, it removes any need for kana here (and it would already have Hepburn as the initial version).

And the Celsius reference was in regards to: "most people have a room temperature IQ".

As a side note, while non-SI measurement may be the stardard for some, it causes complication in unit conversion just as macrons cause problems in translation. ;]

72.145.133.228 19:25, 19 March 2006 (UTC)

"[wāpuro] is very common".

No it isn't. Wāpuro is used almost exclusively by Western fans of anime and manga. Have you ever read any scientific papers or books on Japan and its culture in a Western language? You won't find wāpuro rōmaji used in Monumenta Nipponica or the Japanese Journal of Religious Studies etcetera.

On a side note, even in Celsius the argument "most people have a room temperature IQ" holds true, unfortunatly. JeroenHoek 19:53, 19 March 2006 (UTC)

Looking at the Hepburn romanization page itself, I can see that modified Hepburn is, "...still mainly the preserve of linguists." You're arguing formality, and I'm arguing commonality. Many Westerners are familiar with it, as you point out, and many Japanese are familiar with it through using QWERTY keyboards. It does stand that it is very common, even if you believe that it's not "common enough". It would probably be easy for a Japanese person to learn it if they don't already because it's directly rather transcribing the sounds of the kana.

Those books are used for purposes other than learning how to use Japanese, I believe. If you look at something which a Japanese person would use to learn Japanese, they would learn the kana. Using no Roomaji is even becoming a more popular way to teach Japanese. It re-inforces learning the kana and breaks the connection with English letters, not to mention properly teaching the difference between oo and ou.

If you don't value my opinion, perhaps attention can be paid to the Wikibooks decision, Wikibooks: Japanese Discussion - Resolved Issues. They decided that, when designing for those who know no Japanese (no kana, kanji, nothing), to teach kana right away and avoid using Roomaji. Beyond that, they decided that, when Roomaji was necessary, long vowels would, "... be written according to kana spelling (aa, ii, uu, ee, ei, oo, and ou)".

If we are not interested in proper spelling, why bother with a rough guide to pronunciation which only helps linguists and certain people who are familiar with only Japanese pronunciation using only Hepburn? As discussed earlier, macrons are more likely to be dropped off, leaving us back at the layman version of it (which is fine for them). So why not make Hepburn be the basic spelling (allowing the average person to drop off the occasional ' or macron) while having a highly phonetic version for those who are so interested? Since we already have Hepburn, people who only need that to pronounce it or who just want to write it basically with English characters can get by with that. And people who are ignorant of how to pronounce Japanese can be greatly assisted by the broken apart version (with ', -, and .), and it even shows which sounds go with which kanji. It could (and, in my opinion, should) clear up the ou/oo issue, but if too many people are hard-set against that, kana can clear the issue right up. The pronunciation key thusly helps novices who are struggling and advanced learners who are looking to perfect their knowledge, exactly as any pronunciation key should.

One important question is: why are the kanji included, only for people who wish to write them out by hand or copy-paste them? How easy is it for a person to type them out again when they don't know whether to push u or o (or う or お) when the time comes? You can get entirely different kanji or words if you use the wrong ones. The kanji (and the entire pronunciation key) would be much more helpful if changes were made, and people could actually reproduce it accurately without copy-pasting (whether referring to the macrons or the kanji).

72.145.133.228 21:50, 19 March 2006 (UTC)

Ah, of course. Since an elongated sound is Japanese is really--most technically--two syllables, it would make most sense to have it broken apart thusly:

Jun'ichirō Koizumi (Japanese: 小泉純一郎、 Ko-i'zu'mi Ju'n-i'chi-ro'u, born January 8, 1942)

or perhaps

Jun'ichirō Koizumi (Japanese: 小泉純一郎、 Ko-i'zu'mi Ju'n-i'chi-ro'(u), born January 8, 1942)

72.145.133.228 02:51, 20 March 2006 (UTC)