Wikipedia talk:Manual of Style/Japan-related articles/Archive 1

Hiragana and katakana
Where are the hiragana (&#12402;&#12425;&#12364;&#12394;) characters? These are used more than katakana (&#12459;&#12479;&#12459;&#12490;), which are the only demonstrated set. Katakana is used to write foreign words, onomatopoeic words, or biological names, but for most else it's supposed to be hiragana. I have a chart of kana that I've placed somewhere. When I can dig it up, I'll put up a link to it, for anyone interested. - coldacid 20:00 -0500 2004-03-24


 * The characters themselves are in the charts for Hiragana and Katakana, and the romanization is given in the Hepburn charts. &mdash; Gdr 09:06 2004-04-01

Syllabic n
I have a problem with point 3 in the Romanization section:

"Syllabic n &#12531; followed by b, m; or p is written m."

This is not only not standard practice, it's also confusing and misleading. The syllable "n" exists in Japanese; there's no reason to transcribe it as "m". Exploding Boy 01:59, Apr 1, 2004 (UTC)


 * Also, what has happened to "Naming conventions Japanese"? The link just redirects to this page and there's nothing on it here.  As far as I recall no consensus was ever reached on the issue either. Exploding Boy 02:01, Apr 1, 2004 (UTC)


 * I rescued the material about given name/family name ordering and put it on this page. The only other material was about names of Emperors, about which I know nothing, so that's still waiting for a knowledgeable person to add it. Gdr 15:35, 2004 Apr 3 (UTC)


 * It's still there if you look at the page history. I added a link at the top of this page. -- Taku 02:08, Apr 1, 2004 (UTC)


 * The 'm' romanization is not standard, but it's in widespread use.


 * Here's what Google returns:
 * Edogawa Ranpo:748 vs. Edogawa Rampo: 1630
 * Shinbashi: 17000  vs.  Shimbashi 14600
 * Jinbocho 2420     vs.    Jimbocho 3170
 * ... and most importantly,
 * Monbusho 21,600   vs.    Mombusho 1900


 * And JR uses it: Shimbashi, Nihombashi, etc. - - Paul Richter 02:31, 1 Apr 2004 (UTC)


 * Actually using "m" is a part of standard Hepburn romanization, and the reason for using it is simply that &#26032;&#32862; is pronounced "shimbun". However, revised Hepburn (adopted by the Library of Congress among others) uses "n" in all cases.


 * Personally, I lean towards the revised style, but there are some words where "m" is pretty well entrenched or used in the official romanization, eg. Asahi Shimbun, Namba Station. Jpatokal 03:53, 1 Apr 2004 (UTC)


 * I'm neutral about this; m gives the best indication of pronunciation but n isn't so bad.&mdash; Gdr 09:21 2004-04-01


 * Maybe the right thing to do is to leave this open, with a recommendation to use whatever is most common in use. Gdr 15:50, 2004 Apr 2 (UTC)


 * Howzabout we agree on n for new/obscure/equally popular words, but allow m for entrenched usage? Jpatokal 15:17, 5 Apr 2004 (UTC)


 * There were no loud objections, so now it's policy. Jpatokal 04:20, 10 Apr 2004 (UTC)


 * The new policy seems good to me. I shall undo my Monbush&#333; &rarr; Mombush&#333; change. Gdr 09:13, 2004 Apr 13 (UTC)

Hello, I have an opinion about "n" versus "m". I favor "m" because that is the Japanese pronunciation.

Somebody cited statistics for "Monbusho" versus "Mombusho", but I think they are not the right numbers to give in a debate about Hepburn romanization, since the "n" is Kunreishiki and Mombusho was the Education Ministry, the only major proponent of Kunreishiki.

Adopting the "n" invites readers to pronounce words incorrectly. If we want people to pronounce incorrectly, we might as well adopt Kunreishiki (heaven forfend!). Hepburn, including the "m", at least gets some people to pronounce some words correctly. Given that we have adopted Hepburn, we should use the "m", which continues to be correct pronunciation. The pronunciation Ra-n-po is incorrect; Ra-m-po is correct. Shi-n-bashi is incorrect; Shi-m-bashi is correct. Ji-n-bocho is incorrect; Ji-m-bocho is correct (at least in the consonants). Mo-n-busho is incorrect; Mo-m-busho is correct (again, in the consonants).

The "m" of Hepburn is not "generally deprecated." It is not deprecated in any way within the Hepburn system. It correctly conveys pronunciation. That's why I advocate using it.

Fg2 02:09, 9 Jul 2004 (UTC)


 * You do realize that n becomes m before b and p in English? E.g. "can be" /k&aelig;m'bi/. The only reason it's more noticeable in Japanese is that syllabic n is longer than English syllable-final /n/. If we start using n&#720;, n&#772;, or n&#809; for &#12435; (hopefully only before &#12383;&#34892; and &#12384;&#34892;), maybe it would make enough of a difference. Until then, all things being equal, I'll take the version that's kana-lossless. --Aponar Kestrel 16:57, 2004 Jul 28 (UTC)


 * I agree that n can be pronounced as m before b and p (even though I don't pronounce it that way myself). Perhaps there's some variation in Japanese, as there is in English (if so, it might deserve an article from someone more knowledgeable than I am). But, I would like to approach this from another point of view. The title of this section n is not universally correct. I learned it unquestioningly as syllabic n but the n is often simply incorrect. It's more often correct than syllabic m would be, and far more convenient than syllabic n or m, whichever may be appropriate in the individual case at hand (I'm saying this with tongue in cheek!). The hiragana &#12435; is neither an n nor an m; it changes according to the situation. Now, I know better than to try to get the world to change the name of syllabic n. That's not what I'm trying to do. What I am trying to do is to get people to romanize &#12435; as n when the English pronunciation is n and to romanize it as m when the pronunciation is m. If I knew those pesky symbols of SAMPA or IPA I'd say /n/ and /m/ (and maybe they're what I mean, or maybe not). In either case, it's kana-lossless, since syllabic n and syllabic m represent the same kana, &#12435;. Whew! I wrote more than I planned. Let's hear the counterarguments. Fg2 04:09, Aug 6, 2004 (UTC)&#12288;


 * You're quite right; it is still kana-lossless, and that was stupid of me. That said, I never see anyone write, e.g., "Mombusho" (excepting romanization discussions). The only actual words I can think of where 'm' is commonly used are "shimbun" and "Kampai". Whichever way we do it, we do have to break consistency occasionally -- but we do it far less with 'n' than 'm'.
 * I also think it's a little odd, linguistically speaking, to give only one of &#12435;'s four allophones a different rendering while all three of the others are written the same way. (Yes, four: m&#809;, n&#809;, &#331;&#809;, and... I think the primary/default allophone is something like &#611;&#771; in Tokyo-ben, but don't quote me on it.)
 * Anyway, if it were in common use, I'd say 'go for it', but I think it's probably not going to get used all that much by people looking things up (who are more likely to use wapuro anyway, but that's a whole 'nother argument). And let's face it: we're largely talking about a set of people who can see the word karaoke and read it as carry-okie. Using m instead of n isn't going to make all that big a difference.
 * Incidentally, although English varies on a lot of things, this isn't one of them. You do pronounce it that way (at least in that phrase) -- unless, of course, you're thinking about it while doing so. (And yet, year after year, practically every single student in Linguistics 101 is absolutely certain that they don't.) I suspect, given the similar phonotactic structure of these two cases, that Japanese is the same way; but I can't cite a reference on that, so don't quote me on that either. --Aponar Kestrel 08:55, 2004 Aug 6 (UTC)


 * I've met a lot of people who deny that they're saying things I hear them say, or pronouncing things the way I hear them, so it wouldn't surprise me at all to discover I'm one of them! Fg2 09:56, Aug 6, 2004 (UTC)

Particles
Revth, you edited Japanese grammar changing e to he and o to wo. But Manual of Style for Japan-related articles says to use e and o. Can we come to an agreement? Gdr 15:50, 2004 Apr 2 (UTC)


 * There was no answer, so I reverted the changes. Gdr 08:49, 2004 Apr 13 (UTC)

Double n
Although rule 4 indirectly addresses this, what's the standard for a doubled n? Should "emperor" be tenn&#333; or ten'n&#333;? We seem to be using the former, and I prefer that because there's not really an ambiguity that would require an apostrophe, but computer dictionaries seem to only like the apostrophe version and I do see it in stuff by Viz, so it should probably be mentioned. DopefishJustin 15:46, May 10, 2004 (UTC)


 * Definitely tenn&#333;, writing "ten'n&#333;" is hypercorrect and pointless. I'm also not sure what computer dictionary you're using, as all Linux IMEs that I know of are quite happy without the spurious apostrophe...  Jpatokal 16:32, 10 May 2004 (UTC)


 * I haven't tried entering Japanese on Linux. If you enter tennou in Microsoft's IME, you get *&#12390;&#12435;&#12362;&#12358; (it considers the double n to be the single character &#12435;), so to get &#12390;&#12435;&#12398;&#12358; you need to enter ten'nou or tennnou. In the JquickTrans dictionary (which I usually use), entering tennou gets you *&#12390;&#12387;&#12398;&#12358;, the apostrophe again being necessary. I guess you can explain these away as dumb programming, but why Viz renders &#20096;&#20185;&#20154; as Kamesen'nin is beyond me. DopefishJustin 19:20, May 13, 2004 (UTC)


 * How IME's handle input is not really relevant. What matters is the published form and, as you said, it isn't ambiguous in this case. The basic rule is to use n' when it is ambiguous (i.e. when the n is followed by a vowel or y). Say we have a name like "Sanin".  We can't tell if it is Sa-n-i-n (San'in') or Sa-ni-n (Sanin) unless we have some way to distinguish n from ni. Or kani vs kan'i etc.  Viz is just unnecessarily applying this rule. Mdchachi|Talk 16:19, 14 May 2004 (UTC)

Image captions
It doesn't seem to be possible to use Unicode characters in captions for images, so should long vowels just be unmarked? DopefishJustin 15:46, May 10, 2004 (UTC)


 * Yes, it's the lesser of two evils. Jpatokal 16:32, 10 May 2004 (UTC)


 * New MediaWiki version makes this moot (Unicode now works in image captions), so I removed the note about it. DopefishJustin (&#12539;&#8704;&#12539;) 20:03, 29 May 2004 (UTC)

On Japanese words with established English spellings
I notice in the style manual in the romanization section an example of long vowels is given using the name Tokyo, written as T&#333;ky&#333;. While this is a correct romanization of the long o, Tokyo is almost exclusively written without macrons in an English context. The same applies to other words like sumo (sum&#333;), Kyoto (Ky&#333;to) and Hokkaido (Hokkaid&#333;) which have established English spellings. This should be noted. Exploding Boy 00:55, Jun 26, 2004 (UTC)


 * I removed the examples and added a new section on established English spellings. Please improve. Gdr 14:30, 2004 Jun 26 (UTC)

Google not returning numeric Unicode entities
Apart from the fact that it's a PITA to use numeric entities at all, Google seems not to be returning them, as suggested on this page. this page will do it, though you need to view source, perversely, to see the entities, and I suspect it only handles the Japanese range. (unsigned comment by User:Adamrice)

I agree that it is a pain. Maybe one day the English Wikipedia will transmit in UTF-8 (see Unicode). In the mean time the CGI script below may be useful. Gdr 09:40, 2004 Jul 9 (UTC)

htmlify.cgi #!/usr/local/bin/python import cgi import sys sys.stderr = sys.stdout print 'Content-Type: text/html; charset=utf-8' print print '&lt;html>' print '&lt;head>&lt;title>Convert text to HTML&lt;/title>&lt;/head>' print '&lt;body>&lt;h1>Convert text to HTML&lt;/h1>' def conv(char): o = ord(char) if 32 &lt; o and o &lt; 127: return cgi.escape(char) else: return '&amp;#%d;' % o   form = cgi.FieldStorage if form.has_key('text') and form['text']: t = unicode(form['text'].value, 'utf8') u = '&#39;.join(map(conv, t)) print '&lt;table>' print '&lt;tr>&lt;td>To get this:&lt;/td>&lt;td>' + u + '&lt;/td>&lt;/tr>' print '&lt;tr>&lt;td>Write this:&lt;/td>&lt;td>' + cgi.escape(u) + '&lt;/td>&lt;/tr>' print '&lt;/table>' print '&lt;form action="htmlify.cgi" method="POST">' print '&lt;input type="text" id="text" name="text">' print '&lt;/form>' print '&lt;/body>' print '&lt;/html>'

Arbitrary division 8

 * I don't desire to get into a discussion over the relative merits of WTB's concerns, but I do agree that there needs to be some clarification as to what the style should be. I tend to agree with WTM that we should chose the more "pure" form of the artist's vision-- in the case of some imports, and Viz's in general, that tends to be the manga. (Anime edited for Y7 is hardly ever true to original form.) I don't want to pass judgement, but it seems that WTM is mostly guilty of following (to excess?) the spellings in that "pure" manga, which it turns out is riddled with errors.
 * I'll save Whisper the trouble here: you can't possibly agree with him on that, because that's not what he thinks. His preference is to use whatever localization is used in the manga. To my knowledge, no one here has so far advocated a 'purist' standpoint (although Wikiwikifast briefly did over on Talk:Sailor Moon after WTM moved Usagi Tsukino to "Bunny Tsukino"). I personally prefer the original names, but I accept that I am usually in the minority: and besides, Wikipedia describes what is, not what should be. --Aponar Kestrel
 * I'm thinking about the "populist" standard of article naming (using the Google Test), and I can see the merits of that. While not as pure, it is possibly more approachable to the average end user. My concern with this is two fold, can you tell me how to proceed in these cases:
 * Proper character names. Generally, the first name of a character is more Google-popular than the character's full name, even when narrowed down to the topic at hand. To take your example, "Usagi Tsukino" is less popular than just "Usagi" (when qualified to just sailor moon pages) yet we both agree that just "Usagi (Sailor Moon)" is not the appropiate name for the article.
 * This comes under the general category of 'use the shortest form that doesn't cause confusion'. For example, 'Usagi' could also mean the title character of "Usagi Yojimbo". We probably don't need a disambiguation page titled 'Usagi', because then we'd also have to have one for 'John', or at least 'Sakura'. And so forth. --Aponar Kestrel
 * Inconsistencies of sources. It is odd (to me) that we can have character names from different sources in character lists. We could have (hypothetically) "Tuxedo Kamen", "Sailor Mini-Moon", and "Bunny Tsukino" all in the same list, though all three names are derived from a different source. (Forgive me if I'm wrong, I'm trying to illustrate a point. I haven't watched/read Sailor Moon in years.) This is possibly fine, but is a drawback to the purely populist view.
 * That assumes that you treat each name individually. Personally I think it's more intuitive to treat groups of names like that as a group. However, I suspect that since people tend to seek consistency (internally and externally), this will seldom if ever happen even if one doesn't. --Aponar Kestrel
 * Trends over time. A third issue, and one that is always going to be a problem when you use a translation or popular vote, is that things change. Today, the accepted spelling for a manga not released widely in the US could be one thing. Next year, when the manga reaches our shores in some obscure publication, we could all be updating every article to use those translations. A year from that, we'll be updating again because the anime on Toonami is now driving how people refer to the work. Having a reference source negates some of those difficulties and ensures consistency over time. What has been demonstrated however is that a translation, even with the best of intentions, can rarely be considered a reference source.
 * Such popularity shifts will probably not occur frequently, however. When they are observed to occur -- as they probably will, especially if people document, say, series that haven't been released in the US) -- well, Wiki is not paper; it's easy enough to update, and (if properly written) the articles won't be factually in error even before the update. --Aponar Kestrel
 * If it is appropiate to move these questions elsewhere, that's fine by me. I'm just determining how I can apply these rules in the future and how I should reexamine the articles that I have already contributed. And by raising some of these concerns (playing "devil's advocate"), it might cause us to reexamine or clarify some of the existant policy guidelines. JRP 20:13, 28 Jul 2004 (UTC)
 * No, this is the right place for them. And these are at least new concerns instead of the same-old-same-old rehashed arguments we've been listening to. --Aponar Kestrel 21:25, 2004 Jul 28 (UTC)

I don't know if this is right place to post this question because it involves the right name for an article, but the decision for the right name is made more complicated by Japanese localization problems.

The article in question is Muten-Rôshi (one of the ones renamed by WTM) and a bit more information on this question can be found in that talk page, but it's worth asking here. I'm hoping that knowing the "correct" answer from this example will nail down the rules for me so I can make sure that I use them correctly in the future. (Though, I suspect there is more art here than science.)

The first step would be the Google check, and here's the beginning of that results. (The full table is in the Talk page for the article.)


 * Roshi is a poor name, despite being the most popular by a factor of 3x because it's a generic term (see Buddhist terms and concepts) and we shouldn't override the generic term (even for an article that doesn't exist yet) for a character name. (Note that this name is only the most popular because it catches both the Master and Muten forms of the name. Maybe that makes it not count.)
 * Master Roshi would be a better choice because it adds the familiar qualifier to the generic term. However, that's number three on the list. (Master Muten Roshi doesn't even deserve consideration because I refuse to admit that Viz can honestly on purpose act like that's his real first name, even though they've printed it that way multiple times.)
 * Kamesennin would be a good second choice, but Viz doesn't print it that way (they use Kame-Sen'nin) and it's bad Romaji. (Is it? I'm not the expert.) Do we accept bad Romaji in article titles?

So, given this choice, what's the best pick? The generic name? The bad romaji name? A corrected romaji name that no one uses? The english dub name, even though it's #3 in popularity? Or should I just take the short road and correct the Shonen Jump ô and replace it with &#333;? JRP 22:23, 29 Jul 2004 (UTC)


 * I'm not familiar with the series in question, but diacritics should not be used in article titles. For representing long vowels in Japanese in the content, the sole correct choice (as discussed above in tortuous detail) is &#333;. And yes, "Kame-Sen'nin" is bad romaji; "Kamesennin" would be correct. Jpatokal 00:42, 30 Jul 2004 (UTC)


 * (repeating myself from over on Talk:Muten-Rôshi) You're right about which two to be conflicted on; you should use either "Kamesennin" (which, as Jpatokal has said, is the correct romanization) or "Master Roshi". I suggest that you try to use names drawn from the same source for all the Dragonball Z articles, unless you end up with strongly conflicting majorities on different characters as to which source to draw from -- although I don't think that's likely. If it does happen -- well, just use your best judgment. No worries. --Aponar Kestrel 02:26, 2004 Jul 30 (UTC)

Suffixes
I am wondering does anyone know about any convention regarding suffixes of an article title like prefecture, province and so on. It seems sometimes a suffix is capitalized; e.g. Ise Province, Foo District, Tokushima and some while a suffix is not capitalized like Tokushima prefecture, Tokugawa shogunate. I don't have a preference but I just need to know which one is correct and why. -- Taku 06:37, Aug 1, 2004 (UTC)
 * I don't know. Just today, I was wondering why article titles for prefectures have a lowercase p, and provinces have uppercase. It's very strange to me. Certainly, prefecture and province should be capitalized identically. Wikipedia seems to have (or at one time to have had) a rule that the second and later words should be lowercase, so things that got in early, like prefectures, have the second word in lowercase. When I write, I usually try to put the word Prefecture with the first letter capitalized. It seems much more natural, since it's part of a proper name. Doing so doubles the amount of typing, and I wish Wikipedia had put them in with capital letters to begin with. Possibly, it's a difference between British and American English. Again, I don't know. Sorry I can't be more helpful. Fg2 09:02, Aug 1, 2004 (UTC)

American English always capitalizes place names in full (e.g. Yamaguchi Prefecture, Chiba City, Senri New Town). British English occasionally leaves the "common" part of the place name in lower case, but I've also seen British English capitalize the full name, too... so it seems safe to me to capitalize every word in a proper place name. - Sekicho 15:59, Aug 1, 2004 (UTC)
 * Thanks for the explanation. I agree with your plan. Fg2 12:58, Aug 2, 2004 (UTC)


 * Thanks people. Actually I have remembered that sometime before, I and some people capitalize prefectures but someone reverted it and that was the end. Now, I think we agree to capitalize prefectures. I will put a notice regarding suffixes at the main page. -- Taku 00:17, Aug 3, 2004 (UTC)

Place names
Here's another issue that needs to be resolved: how should articles for places of interest like &#27861;&#38534;&#23546; be named? Here are some options:
 * 1) Horyuji, the plain Japanese (Google: 6,290 but this includes #2)
 * 2) Horyuji Temple, Japanese with a redundant gloss (Google: 1,860)
 * 3) Horyu-ji Temple, disambiguate suffix Lonely Planet style (Google: 748)
 * 4) Horyu Temple, as Taku just suggested on the Manual page (Google: 334)

Personally, I don't like any of these, but I think number 2 is the least of four evils, although number 3 comes close behind since it makes it clearer that "-ji" means temple. Number 1 is basically not English, while the fourth without -ji just sounds strange and isn't popular on Google either; I would say "Horyuji" or "Horyuji Temple" in English, but I'd never say "Horyu Temple". Opinions? Jpatokal 13:10, 4 Aug 2004 (UTC)


 * I think #2 or #3 are conventional. I don't have much preference between them. Another question is whether to capitalize temple. I think it is not a proper noun like Edo period but I am not so sure. -- Taku 13:30, Aug 4, 2004 (UTC)


 * No, "temple" should definitely be capitalized if it refers to a specific temple. "Horyu Temple" is right. Just like Yasukuni Shrine and Osaka Castle. - Sekicho 14:54, Aug 4, 2004 (UTC)


 * Sekicho does have a point there... although "To Temple" and "Senso Temple" still look exceedingly bizarre to me, and as shown earlier Google also favors "Horyuji" form by 6:1.  But if there are no better ideas, can we agree on Horyu Temple (&#27861;&#38534;&#23546; H&#333;ry&#363;ji) as the standard form? I'm itching to do something about the horrible mess that is Category:Buddhist_temples. Jpatokal 07:07, 5 Aug 2004 (UTC)


 * 1-2-3-4 as listed would be my preference in order ... I am not a fan of calling Google searches an authority (see my comments on name order above) but in this case I think they've got it right. I would have no problem with 1, 2, or 3 but as has been mentioned above although 4 might be more "accurate" than 2 or 3, it looks bizarre (and looks bizarre precisely because practically no one would refer to these places as Horyu Temple or Kinkaku Temple in English).  Ironically it's the "bizarre" form that most people seem to call the shrines by (Ise Shrine, Izumo Shrine, etc.)  But if we're just talking about temples I prefer 1 ... it's shortest and most common according to Google and my own experiences (I can't recall, for example, ever hearing English speakers say "Kinkakuji Temple", instead just "Kinkakuji").  2 or 3 adds the redundant "Temple" but it clarifies that the article is about a temple so that serves a purpose.  The hyphen is a matter of asthetics ... personally I would think most people searching for Kinkakuji in Wikipedia aren't going to take the time to type it, but as long as we're consistent that's what matters. CES 07:45, 5 Aug 2004 (UTC)


 * Examples like To Temple and Kiyomizu-dera Temple make it pretty clear to me that we should omit "Temple" and rely on the Japanese (ji, do, in). I see two separate issues here: one is making sure that people can find the article; we can cover that with redirects. The other is internal consistency, and that's most visible in Categories. I changed a couple of articles just now, not knowing that this discussion was going on. (I might have read it and forgotten it --- having lots of things on my mind this week --- or I might not have read it --- not sure. I did read and participate in the discussion about provinces and prefectures.) There's only one left of any significance, Todai-ji. I didn't change that because lots of pages link to it, and furthermore, it seems best to make a decision about whether or not to put the hyphen in. I started a lot of the articles for Kyoto and Nara temples, without the hyphen in most cases, but in truth I don't care, and either way is fine by me. One last comment. I tried several times to change the way the name of Byodoin displays in the category, to distinguish it from Byodoin Temple in Hawaii, making it easier for the reader to select an article. I don't want to change the name of the article, just to make "Kyoto" or "Uji" or "Japan" appear after its name in the Category:Buddhist temples. Nothing worked. Anyone know how to do it? Fg2 08:33, Aug 5, 2004 (UTC)

And now I'm leaning towards option 1 also. Poking around, non-Japanese temples with well-known foreign names (eg. Angkor Wat, Wat Phra Kaew) are also listed without the redundant "Temple" tacked on the end, so there's precedent for this too.

It's tough to make a universal rule though, since as Sekicho says eg. Ise Shrine and Osaka Castle are the usual names, not "Ise Jingu" or "Osakajo". But eg. Meiji Jingu and Izumo Taisha are usually known by their Japanese names. What to do? Jpatokal 09:54, 5 Aug 2004 (UTC)


 * Mention of Meiji Shrine in this context made me laugh. A long time ago, a TV show called The Love Boat did a two-hour special in Japan where the characters visited "The Meiji Jing Shrine." For shrines, I'd suggest using the proper name, followed by "Shrine," instead of jinja, jingu, sha, gu and other suffixes. Thus, Meiji Shrine (as the article is already named), Ise Shrine, Heian Shrine. The spoiler is Toshogu. I've never heard it called "Tosho Shrine." For castles, I think we should use the proper name followed by "Castle" as we have in Edo Castle, Himeji Castle and others. BTW, I think Edo Castle and Kokyo should be merged, being two short articles about the same place. Fg2 22:11, Aug 5, 2004 (UTC)

My vote on this would be 1-2/3-4 (I think the difference between 2 and 3 must be considered on a case-by-case basis, which isn't very helpful). Generally speaking, when English speakers who are knowledgeable about Japan talk about temples and shrines, we say "Ise Jingu" or "Kiyomizu-dera" (and yes, we do say it with exactly that punctuation and capitalization!). "Heian Temple" sounds very odd to my ear: I don't think I've ever heard anyone say that.

The Japan Stylesheet, which was put out by SWET about 15 years ago, recommends that you do use Japanese suffixes and never use redundant English and Japanese (so don't say "Shirakawa River")

The exceptionn is cases where there's already a momentum of English usage in the other direction. For example--Mt Fuji, not Fuji-san (and definitely not Mt. Fuji-san). adamrice


 * Even rivers are a bit tricky I think ... in English it's natural to call Sumidagawa "Sumida River" but calling a river called Shirakawa "Shira River" seems a bit odd. But maybe that's just me? =)  Also, adding the word "River" also helps to distinguish between a river, a place name, and a person named "Shirakawa".  We'll have to tackle rivers next week! =) In any case, would we agree that #1 is the concensus opinion for temples?  Hyphens have their pros and cons ... as long as we're consistent I think that it's not a really big deal.  Any one here have a passionate opinion one way or the other? CES 00:51, 6 Aug 2004 (UTC)


 * For temples, I agree with #1: Horyuji (with or without a hyphen, I don't care much either way). Regarding rivers, we don't have much yet, so agreement based on careful reasoning would be extremely helpful in minimizing the need to move articles. I tend to agree with kawa/gawa: Kandagawa, Edogawa, Arakawa, Tamagawa, Oigawa, Tenryugawa, Chikumagawa, Shinanogawa, Shirakawa, Yodogawa, etc. etc. With these, I fall pretty clearly into the no-hyphen camp. Fg2 03:42, Aug 6, 2004 (UTC)

I see one potential problem with a hyphen. There are some place and personal names containing ji like Kisshoji (place) or Tendouji (person name). One reason to add a suffix temple is disambiguation. But other than this, I am fine with horyuji or horyu-ji -- Taku 15:29, Aug 13, 2004 (UTC)

I like the disambiguation factor too ... also, using the hyphen might make the place names a little easier on the eye for longer temple names. CES 18:34, 26 Aug 2004 (UTC)


 * I don't have a strong preference between "Horyu-ji" and "Horyu-ji Temple", but I do have a very strong preference for the use of a hyphen for all endings (e.g. river, temple, etc), because it clearly marks off the ending from the actual name; that way, with the places where the name happens to have the same ending (e.g. the examples above by Taku), naive readers will know it's actually part of the name. Noel (talk) 15:45, 2 Nov 2004 (UTC)


 * I am not a big fan of the hyphen, especially in the case of those temples where the 'temple' ending is commonly attached in English. (e.g. Byodoin, Todaiji, Kiyomizudera, Hachimangu) But, that said, my preferences lie in the order presented above (1-2-3-4). Using the regular Japanese name is most preferable because that's the actual name of the place. I don't want to see my hometown's sites rendered as "Nueva York" or "自由の女神像", as those are quite simply not their names. Once someone (perhaps myself) gets around to making an article on Buddhist temples in Japan, a user will be able to find explanations of the meaning of -ji, -gu, etc with a minimum of effort. In the end, I'd be happy with either of the top two, but please please don't end up using the fourth option. Horyu Temple, To Temple, Senso Temple, just sound wrong. LordAmeth 11:57, 7 September 2005 (UTC)


 * I just noticed that User:Jpatokal has moved several of my pages (and probably quite a number of others) to include macrons in the article title. Was there consensus on this? I'm personally a big fan of not using macrons or accent marks in the article titles as it makes them less easily searchable, and arguably harder to pronounce or understand for those who are not familiar with the given language. (For example, I know how to pronounce "Curacao"; I have no idea how to pronounce "Curaçao".) See Naming conventions (use English). LordAmeth 11:07, 23 October 2005 (UTC)

Asian languages naming table
Since Wikipedians who mess with Korean and Chinese related articles made naming tables, I decided to lump in Chinese and Korean together to create an "Asian languages" table. In several articles, the "Japanese" got lumped in as well. See: Chopsticks, Go (board game), Liancourt Rocks, Manchukuo, and Sea of Japan as examples. (The Liancourt Rocks table was started by another person who had the same idea.)

NOTE: The "Hepburn" added to the "Romaji" indicates only that the standard way of writing it conforms only to Hepburn - It does not imply that more romanization systems should be added to the table. WhisperToMe 21:26, 28 Aug 2004 (UTC)


 * I may be in the minority on this one, but while I think the naming table is a neat idea, I wonder if it clutters up the articles more than it helps. Some of the articles now have large blanks at the top of the article (Go) while others make you wonder if the naming table will soon be longer than the article itself (Sea of Japan) ... this is probably the wrong forum for a discussion on these tables (since more than Japan-related articles are in question), but I am slightly concerned that with these tables Wikipedia is turning into a foreign language dictionary instead of an encyclopedia. CES 15:01, 31 Aug 2004 (UTC)


 * I'm in agreement with CES on this one, really. In the case of 'Sea of Japan', especially, I don't think it's needed at all: there should of course be a section on the naming dispute, but simply adding something like "Russian and Chinese maps also typically use Sea of Japan (&#26085;&#26412;&#28023; Rìb&#283;nh&#462;i and &#1071;&#1087;&#1086;&#769;&#1085;&#1089;&#1082;&#1086;&#1077; &#1084;&#1086;&#769;&#1088;&#1077; Yaponskoye More(?) respectively)" to the "naming dispute" paragraph would probably be enough to make the box almost useless. More generally, when a box such as this suffices, the names of things or places from multiple languages are almost certainly not notable enough to be encyclopedic. Contrariwise, when they are encyclopedic, a little box isn't enough; we need a full paragraph of text -- at which point we probably don't need a box anymore, since it's redundant. --Aponar Kestrel (talk) 21:22, 2004 Aug 31 (UTC)

Actually, the naming dispute has its own article (Dispute over the name of the Sea of Japan), and the names and stuff came from that article. WhisperToMe 00:15, 1 Sep 2004 (UTC)


 * ... which means you're duplicating information that already exists somewhere else. That's probably not a good thing. --Aponar Kestrel (talk) 04:22, 2004 Sep 2 (UTC)


 * The reason the info was duplicated is to simply show the Korean/Chinese/Japanese on the page about the Sea of Japan, and then to go into detail on it on the debate over it. All Korean/Chinese articles are supposed to show the Korean/Chinese characters and romanizations of them. WhisperToMe 06:33, 5 Sep 2004 (UTC)


 * I'm in agreement with the two people above. Just take a look at the chopsticks article and compare the box with the paragraph below it.  They both convey the same information: the box takes up more space and distracts from the actual article. - Sekicho 01:49, Sep 1, 2004 (UTC)


 * The reason the boxes were made is because Korean and Chinese have two largely-used romanization systems (as opposed to Japanese only having one), and that Korean also has the Hanja and Hangul. In my opinion, a paragraph with the right information can do more than a table. But unless specific information on its usage can be found, a table is better. WhisperToMe 02:39, 1 Sep 2004 (UTC)
 * But if specific information on its usage can't be found, is it really all that interesting? That said, there are a very few cases where a table would be useful. May I suggest Zen? --Aponar Kestrel (talk) 04:22, 2004 Sep 2 (UTC)

Romanization redux
Whisper has pointed out that some of the Sailor Moon characters have strikingly common nonstandard romanizations. Of course the situation is even worse for &#22825;&#29579;&#12399;&#12427;&#12363; Ten'&#333; Haruka. I converted these to "Meioh Setsuna" (etc.) in the article text, but I'm wondering if this was really the right thing to do here. I could make a strong case that Meioh should be treated as no more worthy of use than Mei&ocirc;, but I don't know what the "average reader" would be least surprised by in this circumstance (especially given that any group of five readers will likely be split as to their preferred spelling). So, generally speaking -- do people agree with my first impulse ("Meioh": vox populi, vox Wikipedias) or my second ("Mei&#333;": damn the otaku, full speed ahead)? --Aponar Kestrel (talk) 23:17, 2004 Aug 31 (UTC)
 * Canonical form: &#20901;&#29579; &#12379;&#12388;&#12394; Mei&#333; Setsuna
 * Google search for "Meioh Setsuna" (14,000)
 * Google search for "Meiou Setsuna" (870)
 * Google search for "Meio Setsuna" (92)
 * Google search for "Setsuna Meioh" (4,910)
 * Google search for "Setsuna Meiou" (734)
 * Google search for "Setsuna Meio" (3,400)
 * For me, it depends on why the naming is used more commonly. If it is because official publications spell it that way (As in the case of the Japanese and English versions of the Sailor Moon manga, which use "Kaioh", "Meioh", and "Tenoh"), then I would do so on an article. If it is because fans like using wapuro, even though the official publications don't do it, then I would default to the official spelling, or if none exists, the MOS Style spelling. WhisperToMe 01:38, 1 Sep 2004 (UTC)
 * Ten'&#333; vs. Tenn&#333; discussion moved to Talk:Haruka Teno. --Aponar Kestrel (talk) 07:40, 2004 Sep 2 (UTC)

Another way to show a macron
Someone found another way to write a macron. He or she used a code to get o. The macron is "higher" than the one in the standard character (&#333;). I did not know about that code until seeing it in the Lupin article. WhisperToMe 06:27, 5 Sep 2004 (UTC)
 * I put a message on that user's talk page saying how to get the usual macrons. Someone once scolded me for using undefined to get superscripts. Something about a limit on the number of something-or-others that can go in an article. Leaving aside limitations, I don't know why the higher bar might be better... maybe it's easier to see?Fg2 06:51, Sep 5, 2004 (UTC)
 * I believe that many older computers may not be able to see the traditional "macron" character. WhisperToMe 16:55, 5 Sep 2004 (UTC)
 * Using isn't quite wikicode to get a macron. Instead, it references Template:o and includes it inline. This is a bad thing because then the Wikipedia server has to go and fetch another file and perform the inline substitution. To prevent possible denial-of-service attacks based on Template: abuse, only five templates are permitted per article; thus frequent use of  and similar constructs will cause templates to break. Please continue to use &amp;#333; et al. --Aponar Kestrel  (talk) 17:22, 2004 Sep 5 (UTC)
 * ... I should add that older computers won't understand " " either, so that's no argument at all, I'm afraid. --Aponar Kestrel (talk) 17:58, 2004 Sep 5 (UTC)
 * Okay. I let the person know on his talk page that he can't use the template to get a macron. :) WhisperToMe 18:26, 5 Sep 2004 (UTC)
 * The five-include limit is brain-damaged, which I complained very loudly about until somebody finally reassured me that it's been fixed, though the fix will not go live until MediaWiki 1.4. I still think templates are the way to go for this, but not until the restriction is lifted again.
 * "brain-damaged"? That's a little strong, Furrykef. Anyway, I've gone in and edited Template:A, Template:O, and Template:U -- apparently User:Kukkurovaca created them for ease of use in transliterating Sanskrit. Ick. The less stress on the server, the better. --Aponar Kestrel (talk) 19:38, 2004 Sep 5 (UTC)
 * It may be strong, but it's not literal; no offense to those who implemented it, but it really is not the Right Thing. Moreover, I doubt this use of templates creates any special stress on the server, as rarely as these templates are likely accessed. - Furrykef 02:54, 6 Sep 2004 (UTC)
 * If there is going to be a vote on this, I would go with using the HTML character reference. :) WhisperToMe 02:58, 6 Sep 2004 (UTC)

Never mind that isn't actually the correct character... Exploding Boy 15:08, Sep 7, 2004 (UTC)
 * It used to be. - Furrykef 02:29, 8 Sep 2004 (UTC)

I doubt that, but it's not any more. Exploding Boy 21:00, Sep 8, 2004 (UTC)

These templates have all been deleted now - the Insert: box on edit page now provides a way to insert most special characters. Noel (talk) 15:39, 17 Mar 2005 (UTC)

Use of given names of historical figures
This was moved from Tokugawa Ieyasu.

I changed "Tokugawa" to "Ieyasu" in many places because that distinguishes Ieyasu from his descendants. Likewise, "Toyotomi" --> "Hideyoshi" and "Oda" --> "Nobunaga".


 * I dunno. This seems like a bad thing to me.  People who don't know better are going to start thinking that Toyotomi and Oda are their given names, not their family names.  Is there really a need to distinguish these individuals from their descendants? Sekicho 12:35, Aug 27, 2004 (UTC)


 * Yes, I think so. Nobunaga's son died with him, but he had a brother, Yurakusai (if I remember correctly). Hideyoshi's son is mentioned in the article together with Hideyoshi, so saying "Toyotomi" seems unclear to me. Likewise, Ieyasu's sons figure in this article, so it seems necessary again. I looked at the articles on Nobunaga and Hideyoshi, and they prefer to use the given names. (However, I didn't look to see whether or not I'd changed them --- I might have!) Similarly, British kings go by given name: James, Charles, George etc. rather than by family names (Windsor, and so forth). I know that these generals weren't kings or emperors, but perhaps it's by that analogy that they're commonly called by their given names in English. After all, they wanted their positions to be hereditary. This makes them different from modern military generals in the US or Britain. Anyhow, it's not the most important thing. If there's a consensus to go back, I'm ok with that. Fg2 12:50, Aug 27, 2004 (UTC)

I think Fg2 is correct. It it just a convention to use given names instead of family names in articles like this. I looked at an article in Britannica and it uses Nobunaga:


 * Meanwhile, Nobunaga promoted a new economic policy by abolishing the collection of tolls on the roads and from the guilds, both of which had been privileged sources of income for the local daimyo.

Coincidentally, they got really lengthy and balanced articles of Japanese historical figures. We are still far behind. -- Taku 17:47, Aug 27, 2004 (UTC)


 * I would support using given names if we were writing about several Tokugawas in one article. I don't think we should use given names all the time, especially when there's no chance of confusion (such as this article, where "Oda" would obviously refer to Oda Nobunaga.) - Sekicho 23:57, Sep 5, 2004 (UTC)


 * I guess Oda's pretty clear in this article!Fg2 07:00, Sep 6, 2004 (UTC)


 * I don't think this is an issue of ambiguity. It is just so common in writing Japanese history in English to use Nobunaga instead of Oda as I quoted from Britannica. -- Taku 02:43, Sep 18, 2004 (UTC)


 * There is an issue of ambiguity, of course: if it might not be clear which of two figures with the same surname are meant, then we need to use the given name. Likewise, in the relatively unlikely event that we're discussing two people with the same given name, we need to use the surname. However, if neither would be ambiguous, then as far as I'm concerned this is no more important than British vs. American spelling, and I don't think it's worth the trouble of fixing (except for the sake of consistency within a given article). --Aponar Kestrel (talk) 01:47, 2004 Sep 20 (UTC)


 * Second Aponar Kestrel. This needn't develop into a policy, but I think it's perfectly reasonable (and attested to elsewhere) to use given names in this context. Now, if there are articles about the history of Kabuki, and we need to refer to all the different Ichikawa Danjuros out there, we've got another problem. adamrice 13:27, 20 Sep 2004 (UTC)


 * There is no problem there. They are always referred to with the suffix Roman numeral, e.g. Ichikawa Danjuro IX, and the articles should be titled that way; if you need a short name for them, refer to them as Danjuro IX - that's the usual practise in the Western writings on kabuki which I have. Noel (talk) 16:00, 2 Nov 2004 (UTC)

I don't think we should develop a convention, which is not seen anywhere but here. As I said, my point is we should follow what almost all history books do. Sometimes the names Oda and Toyotomi are unambigous, but it is just too normal and accepted practice to use given names refer to historical figures. -- Taku 16:00, Sep 20, 2004 (UTC)


 * I concur. In all my history books, when they are referred to with only one name, it's always as Hideyoshi, Nobunaga, etc. (Perhaps this is because, at least in Hideyoshi's case, his son is an important actor in the closing days of Ieyasu's consolidation of power, so Toyotomi would be ambiguous, and of course the full name is clunky.) Noel (talk) 16:00, 2 Nov 2004 (UTC)

I don't think we should develop any convention either. In the English version of the manga Naruto, the character's names are in Japanese order, and they always refer to each other in short hand by their given name, perhaps as a way of reflecting what was done in historical periods of Japan. WhisperToMe 22:54, 20 Sep 2004 (UTC)


 * Not everything about Japan and Japanese is reducible to manga. Exploding Boy 18:39, Sep 21, 2004 (UTC)


 * No, for that, we need &#x7537;&#x306f;&#x3064;&#x3089;&#x3044;&#x3088;&#xff01; adamrice 20:50, 21 Sep 2004 (UTC)


 * Furthermore, informal conversation is very different from encyclopedic work (of course people use given names in conversation... formal writing is much different). But I'll accept Taku's argument, although begrudgingly. - Sekicho 20:22, Sep 21, 2004 (UTC)


 * As far as history studies in Japan, it is quite common to use (only) given name for historical figures, at least in Japanese. How about refering them first with their full name and in following only their given names? --Aphaea 00:58, 27 Sep 2004 (UTC)

Random question
... I'm sorry, I don't know where better to ask this. Is there a name for abbreviations like w&#257;puro or Kimutaku? We could really use an article on it, but I don't know what to name it other than 'Japanese abbreviations'. --Aponar Kestrel (talk) 02:27, 2004 Sep 30 (UTC)
 * Check out Portmanteau. It covers combinations of complete words (like portmanteau) as well as parts of words (such as smog). I don't know a specific term for words from Japanese. Often, they're closely analogous to acronyms and initialisms, formed by taking the first kanji from key words in a phrase or name just as acronyms are formed with the first letters (although neither of your examples falls into this category). If nobody has a better suggestion, why not go with "Japanese abbreviations.' Fg2 03:44, Sep 30, 2004 (UTC)


 * Maybe wasei-eigo. Kappa 08:47, 4 Dec 2004 (UTC)

a new page for naming order issue
I have created a separate article about naming order issue: Manual of Style for Japan-related articles/Naming order. Personally, I hate to have a separate page about the same topic and I wish I didn't interrupt any discussion we have been having so far. But it should be helpful for us to summarize discussion and really go to come up with a solution in the near future. -- Taku 04:57, Oct 1, 2004 (UTC)

Rendering of Japanese Names
After looking through some articles on Japan, I got the feeling that we should try to simplify the way Japanese names are given within articles. A lot of articles give names the following way:  (, ) -- e.g. "Kazuhiko Kat&#333; (&#21152;&#34276;&#19968;&#24422; Kat&#333; Kazuhiko)" -- see the article on Monkey Punch. It seems a bit silly and redundant to me to have the r&#333;maji repetition of the name (only in reversed order). I am aware of course that the part following the kanji is supposed to be the transliteration, while the part before the bracket is supposed to be the form of the name commonly used in English, so there is a conceptual difference. But practically, this only means repeating the name and creating unnecessary work. There seem to be the following ways to go about it: No. 1 would seem to include all the information pertinent to a name, but seems to me too much work, unless the English name is very common. No.4 saves a lot of work, but I for one would prefer to see kanji next to names newly introduced in an article, without having to go to a second page. No. 3 would display the name in a format familiar to Western readers -- but is this necessary? If people are unsure about whether something is the first or the last name, they can also look it up on the manual of style. Also, the word order wouldn't match the kanji. My personal favorite would be option No. 2. Obviously, this would be a bit awkward for names that are frequently mentioned in the Japanese media, but if the person is really a celeb, the name would have a link to a page that would also provide the name commonly used in English. Any thoughts? Perseus 05:37, 1 Oct 2004 (UTC)
 * 1) drop all macrons from the first mentioning of the name, thus creating something like an "English rendering" of the Japanese name (thus: "Kazuhiko Kato (&#21152;&#34276;&#19968;&#24422; Kat&#333; Kazuhiko)")
 * 2) only use the r&#333;maji rendering, in last-name first-name order  ("Kat&#333; Kazuhiko &#21152;&#34276;&#19968;&#24422;")
 * 3) use the r&#333;maji rendering, in first name-last name order (Kazuhiko Kat&#333; &#21152;&#34276;&#19968;&#24422; )
 * 4) drop both the kanji and the r&#333;maji rending if the name has a link (so that people interested in the kanji / transliteration can look them up -- e.g. Kazuhiko Kato); if there is no link, proceed as until now.


 * I would support number 2, but I think some people might have some objections, especially with the naming order. (See Manual of Style for Japan-related articles/Naming order for more information.) [[User:JoshG|Josh | Talk]] 00:45, Oct 20, 2004 (UTC)
 * And yes, I realize that this idea kind of conflicts with my statement in the naming order discussion. [[User:JoshG|Josh | Talk]] 14:39, Oct 20, 2004 (UTC)


 * I prefer a combo of 1 and 4 - 4 if there is a link, 1 (with full romaji with macrons, etc in Japanese order, in parens immediately following) if not. Noel (talk) 16:05, 2 Nov 2004 (UTC)