Wikipedia talk:Manual of Style/Japan-related articles/Archive 3

City names

Discussion continued from Wikipedia talk:Manual of Style for Japan-related articles/Cities

I am not so sure about what I might accomplish here but I want to try to write about some historical background and sort out things. (I have been following the debate but has been distracted by a snap election in Japan :) I was hoping to see some consensus to emerge but it didn't happen very unfortunately.

Among issues we have here, one is about the procedure; that's how to come up with a convention and enforce it. I am the who started this page with a hope to ensure the consistency in style and, more importantly, gives a quick guideline to those who contribute to Japan-related articles. In other words, I believe the manual of style must both be consistent about style and reflect the views of the majority of contributors. I don't suppose that there is a correct style and an incorrect one but one that is preferred by many and and one that is preferred by few. To reflect the preferences of contributors, we, however, should look at not just the sheer numbers of contributors but those who do actual many contributions to many articles. In this case, the likes of Rick Block do a lot of tedious but very invaluable edits, and we must agree that we can't ignore the voices of those and the manual of style, if any, needs to be something preferred by such people. They may appear to be authority figure policing articles, but that's something we badly need in wikipedia. I have seen so many valuable contributors leaving wikipedia because they think the editing process here is not functioning, and we all should agree that

In any rate, we are probably all aware of problems so I don't have to repeat here. Instead, I started a new page Explanatory note for Japan-related articles. Some people (i.e., User:WhisperToMe and Jefu) have suggested to put a footnote at each article to inform readers about the conventions. So I thought we can put a link to this page at each article in the form of footnote, and we can prevent disparity from arising between those who follow this manual of style and those who are unaware of it and obliviously create inconsistency when they try to fix a problem. -- Taku 01:11, August 26, 2005 (UTC)

Sounds like a good idea. I think part of the problem is, as you said, many people are unaware of the conventions and why they arose. I commented a little more on the footnote idea below. CES 02:26, 26 August 2005 (UTC)

Hm.. I thought I made an edit here, but it appears not. Resuscitating the tail of the discussion for the moment, can be packed away again if anything is actually resolved.

I disagree strongly that this is a case of needing to 'educate the ignorant' - this is a case where editors focused on the specialist need to learn from common useage. As I see it, editors had a problem with not having a naming scheme for Japanese cities, and came up with a nice simple rule. But when they met valid objections on various talk pages for requested moves, rather than actually trying to rethink the idea and gain concensus on a revision, instead held a quick poll in this tucked-away corner of wikip and then used that as an excuse to railroad the proposal through.

Personally I think the current wording of the MoS (J) is overly prescriptive, and fails to acknowledge that there are always difficult cases and exceptions where rigidly applying a set rule is detrimental. Note how the 'Person names' section addresses the fact there is not a complete consensus on the issue, and links to the relevant discussion. I'd suggest that the placenames section needs to be changed *now* to something similar - by someone who has the diploma or whatever that's required to be allowed to edit the page without just being unthinkingly reverted. At the moment this section just doesn't reflect either consensus or the reality of current (apparent) policy. -- zippedmartin 21:57, 6 September 2005 (UTC)

Japanese Dates

Would a mention of the Japanese system of naming years be useful? I just found an example of some one refering to "Showa 33" as "Hirohito 33" - [[1]]. 07:06, 6 August 2005 (UTC)

A short paragraph with a link to Japanese era name would suffice I think. JeroenHoek 11:35, 6 August 2005 (UTC)

---

Now for my next pet topic: dates. This will only be of interest to people interested in pre-Meiji Japanese history, but I'm noticing a lot of date problems, misunderstandings and inconsistencies in Wikipedia. As many of you probably know, Japan used a lunisolar calendar prior to January 1, 1873. I think we should adopt a standard of primarily using Japanese months and dates for dates prior to this. The reasons are:

1. Virtually all Japanese language history sources use Japanese dates exclusively. In fact I don't think I've ever seen one that has taken the trouble to actually convert them into Western dates, because there is no need to do so.
2. Although I'm not as familiar with English language Japanese history sources, I'm pretty sure that using Japanese dates is the standard in English sources as well.
3. Note that this has no effect on dates after January 1, 1873 (Gregorian), which is the date that Japan adopted the Gregorian calendar, so dates from modern Japanese history would remain as is (this is also the standard in Japanese history books.)

In any event, this has to be a one or the other proposition. If we end up with some articles that use Japanese dates and some that use converted Gregorian or Julian dates prior to 1873, we will end up with a confused mess when trying to compare when various Japanese events happened in relation to one another.

The only exception I would propose is events that are added to the "On this day in history" pages. For events in Japanese history prior to 1873 that are added to these pages, I think they should be converted to Western dates. I've actually been adding a number of such items, and what I have been doing is converting pre-1873 dates to Gregorian, adding the Japanese historical event to that particular date in Wikipedia, and then including a parenthetical at the end of the entry that gives the date according to the traditional Japanese calendar (to avoid confusion.)

Assuming this is acceptable, the only problem is what we should do about the Wikipedia rule on dates that other calendars can be used, so long as the date is given in Julian and/or Gregorian as well (See Wikipedia:Manual of Style (dates and numbers)#Different Calendars. Do people think this is really necessary? If so, what format would you suggest for doing so? One way to do it is to footnote the Japanese date and include the Gregorian or Julian equivalent in the footnote. In that case I think the Japanese date should not be linked (since it would link to the wrong date anyway) but the date in the footnote should be linked. For an example, please see: Emperor Konin of Japan. Any thoughts? -Jefu 07:44, August 17, 2005 (UTC)

Western dates have an obvious advantage in that you can subtract them. It's hard to guess how long the Onin War was when you know that it ran from Onin 1 to Bunmei 9. Knowing that the years were 1467 and 1477 makes it easier. Fg2 12:09, August 17, 2005 (UTC)

Sorry, I wasn't very clear. I'm talking only about the months and days. I do not mean to suggest that we should use Japanese eras for the years. For example, January 18, 1467 by the Japanese calendar (the date that fighting between Hatayama Yoshinari and Hatayama Masanaga broke out, which escalated into the Onin War) was actually February 13, 1467 by the Julian calendar (which is the one Wikipedia mandates since the date was before October 15, 1582.)

It seems like having both dates would be best ... I think either way you have the potential for confusion as some sources will use Japanese dates and others will use Western dates (and the sources might not make it clear which they are using). Wherever possible, I'd suggest having both dates, with the Western date given priority (although the articles are on Japanese subjects, this is still the English Wikipedia). CES 13:45, 17 August 2005 (UTC)

When you say "Japanese subjects," you are using Japanese to refer to a culture/country. But when you say "English Wikipedia" you are using English to refer to a language. In other words, I don't think this has anything to do with the language. I think it has everything to do with the culture. While I don't necessarily disagree that we should probably have both dates (and that is what the Wikipedia style requires), I do not necessarily agree that priority should be given to Western dates. I also note the following from the preface to Volume 1 of The Cambridge History of Japan: "Years recorded in Japanese eras (nengō) are converted to years by the Western calendar, but months and days are not." This is exactly what I am proposing, except that we should probably also include the Western dates (Gregorian or Julian as the case may be) in footnotes, which goes a step beyond The Cambridge History of Japan. -Jefu 15:15, August 17, 2005 (UTC)

What would be the problem with using both sets of dates? john k 15:33, 17 August 2005 (UTC)

Hoo, boy, this is a tricky one. Not sure quite what to do...

While I'm pondering, a few comments on the language/culture issue. For the vast majority of languages around the world, language and culture are deeply connected. That's not the case for English (and there are a few others where this is true, like Spanish). Still, English is the primary language (i.e. language spoken at home by the majority of people) in only a few sizeable countries (UK, US, Canada, NZ, Australia) , and they all have fairly similar cultures. So it is cultural, too. Noel (talk) 13:30, 18 August 2005 (UTC)

Not a problem really, just a bit of a pain to calculate since virtually all sources that give specific dates give Japanese dates only. What I would really like to do with this dialog is come up with a standard format for presenting dates and add a section in this article explaining whatever format we come up with. An example of my proposal can be seen in the article Emperor Konin of Japan. I think footnoting is far less clumsy than putting two dates in the actual text along with labels for which one is which. But I'm certainly open to other suggestions. -Jefu 15:41, August 17, 2005 (UTC)

Hmm...I have to say that it seems really weird having dates not line up like that, and with it only noted in footnotes. I think if people see a date, they will assume it is the Gregorian or Julian date. At any rate, events of significance in both Japanese and western history (like Perry's visit), should certainly give both dates. john k 15:56, 17 August 2005 (UTC)

I looked at two articles on the Japanese Wikipedia. ja:大政奉還 gives dates using both systems and ja:大化の改新 doesn't specify (but the date they gave links to ja:6月12日 (旧暦) which is the old system). Not quite sure where this leaves us... Fg2 05:52, August 18, 2005 (UTC)

Part of the question is what is typical in the West when speaking about Japanese history. I've got to believe that using Japanese dates without conversion is the norm. I don't have that many English language resources on Japanese history prior to 1873 (partly because there simply aren't very many), but I did check two huge standards. Sansom's books seem not to give specific dates for events, only years. One specific date I noticed while leafing through them was the date that the Treaty of Amity was signed with the United States, which was translated into the Western date (which, in hat context, I think makes perfect sense, but for most of Japanese history, it probably does not.) What's the point in giving the primary date of Taika no Kaishin as July 10, 645 when virtually everyone familiar with this event knows it as June 12? And my second source was the Cambridge History of Japan which, as I mention above, uses the Japanese dates as is and only translates the year into Western years (and converting the year is common practice among Japanese historians as well.) The problem with giving primarily Western dates, other than the fact that for 95% of the events in Japanese history there is simply no need to do so, is that the resources out there that people are liable to check are overwhelmingly Japanese and they of course use Japanese dates. If there were tons of English language books on Japanese ancient history that took the trouble to translate all the dates into Gregorian or Julian, I might understand the preference. But as long as we footnote the dates and make it clear that they are Japanese dates that are standard for use in speaking about Japanese history, I think we should be fine. I honestly think translating the dates and giving the western dates will introduce confusion rather than quell it. -Jefu 06:38, August 18, 2005 (UTC)

Someone else, who may join the discussion here shortly, seems to have an issue with using January, February, etc. when talking about months in the old Japanese calendar. I'm not at all animated about this issue. I think as long as you make it clear that we are talking about Japanese dates in the footnote, who cares if you translate ichigatsu as January? The calendars don't line up anyway. Although using phrases for Japanese dates of, for example, "the 12th day of the sixth month, 645" may provide a stronger signal than a footnote that the dates are Japanese and not Western (to respond to a concern raised above), you would then have the problem of what to do about the leap months (known as uruuzuki in Japanese). In the Japanese calendar there were essentially repeated months every few years to keep the months basically aligned with the seasons. So in a year where the repeated month is february, you would have ichigatsu (first month, or January), nigatsu (second month, or February), uruu nigatsu (leap second month, or Leap February), etc. But if you start translating months into english using ordinals you get into a weird situation where it looks like you are using ordinals, but where the leap month (which is the third ordinal month) wouldn't be referred to as the third month. I think it just gets too bizarre. To me it is far more preferable to use the month labels that English speakers are already familiar with and just append "leap" to the uruuzuki. So in my example above you would have January, February, Leap February, March...December. Anyway, we'll let him speak up on this issue when he joins the conversation. -Jefu 06:38, August 18, 2005 (UTC)

I am absolutely, utterly, opposed to using "January" instead of "first month". For one, it's simply wrong (and mis-leading to our readers); the two system are just different, end of story. Second, a lot of the time the months don't even approximately match up, because the lunar New Year often fell a long way from January 1 (e.g. in Meiji 5, when they converted, the 2nd day, 12th month was December 31st; I see one source that says "The first lunar month of the year always corresponded with the sign Pisces during the months of February and March"). Add in intercalary months, and it gets even worse. Either convert the dates properly (see my note below), or leave them in the Japanese calendar. Noel (talk) 23:09, 18 August 2005 (UTC)

I'm slowly beginning to see how the Japanese convention makes sense. If we do make one (Japanese dates) the standard, I would advocate allowing editors to add the other (Western dates) at their option. And an article on the traditional Japanese system of dates, with content like what you just wrote, would be helpful for readers. (It could go in Japanese calendar, but that article's already pretty far-reaching; certainly, a link from there would be useful.) Fg2 07:08, August 18, 2005 (UTC)

My proposal, which I would suggest we add to this Manual of Style, is that authors use Japanese dates prior to 1873 (unless, of course, you don't know it for some reason, but I doubt that would happen) and footnote it. The footnote would say that it is a Japanese date and not a western date, and it would give the Western date, which thanks to several online sites, are relatively easy to convert. I don't think I would advocate using Japanese dates without indicating in some manner that it is, in fact, a Japanese date. For an example of the style I propose see Emperor Konin of Japan

And I would be happy to write an article, or add to the existing one, about the old Japanese calendar. —Preceding unsigned comment added by Jefu (talk • contribs) 07:34, 18 August 2005

Can you please point us to these several online sites? Thanks! Noel (talk) 23:09, 18 August 2005 (UTC)

If most dates in source books about Japanese history are given with the Japanese date, but if this date is indeed easy to convert as the above user indicates (please sign your posts!), why not use both in the article then? This is the English wikipedia, and by English I of course refer not only to the language but the culture of English speakers, most of whom assume a date is in the Julian/Gregorian system when they see a date. Leaving out the western date would be inappropriate in my opinion. CES 12:52, 18 August 2005 (UTC)

Also, I was wondering: is there a reason why English-language sources of Japanese history use the Japanese date? Is that just a convention that arose from translating Japanese sources? No matter which way we end up going with this, we'll need a footnote. CES 13:10, 18 August 2005 (UTC)

Wow, I don't know. It's going to be one heck of a lot of work to convert Japanese dates, isn't it? (Which may be a good part of why nobody does it! :-) Is there a table somewhere which says which months were short (29 days) and which were long (30 days) in each year? (Since the pattern changed every year, with no two years ever using the same pattern, at least in theory.) I seem to recall seeing one in some reference book, but I can't recall where. (I know that in the Edo period e-goyomi were produced as a way of propogating that information, because of the government monopoly on printing of calendars.) And then there are the inter-calary months, but tables of years with those are common.

One thing I'm not too sure about is the start date (in the Western calendar) for each Japanese year; is that easy to work out (based on whether or not the previous year had an intercalary month, etc), or was is based on astronomical events (which would of course require more research)? Actually, a (large :-) table which simple allowed on to look up a Japanese year, and see on which Western dates each Japanese month of that year started would be the best of all, but does one exist? Without some practical means of conversion, the issue of whether we should convert them may be moot. Noel (talk) 13:30, 18 August 2005 (UTC)

I located that table of long/short months; it's in Egoyomi and Surimono, by Matthi Forrer - but it only covers dates from Genroku forward. (I scanned in the first page here if anyone wants to see what it looks like.) We ought to have this info in Wikipedia, so if there's no online source, I volunteer to scan the table in and OCR it (although I'd like a hand for proofreading, etc).

Before I do any more on this, though, can I ask Jefu to point us to those online sources that do date conversion? No sense in duplicating anything... Noel (talk) 23:21, 18 August 2005 (UTC)

I understand it might be a fag, but I think editors should try and convert dates as they write. It's silly to expect readers (of English, primarily familiar with the western calendar) to sidetrack and read an article on Japanese dating systems just to be able to learn about the history of Japan, then perform on-the-fly date conversion, for each and every article on a pre-Meiji topic. Also, I like the wiki feature of having clickable dates in history articles that give context - currently these pages tend to be quite euro-centric, but that won't change unless there's date conversion from other systems. I agree a page explaining conversion is needed, either way. --zippedmartin 23:46, 18 August 2005 (UTC)

As to Noel's comments above, I don't understand the opposition to using the names of dates that are already familiar to English readers. What does "February" convey to an English speaker other than "second month of the year"? Do you think that when a reader sees "second month" they don't automatically generate an association with the word "February" anyway? To give a perfect example of how this does not create confusion, I give you the Japanese language. The second month of the year is called nigatsu whether one is speaking about the old calendar or the new. And although nigatsu is literally second month. Japanese readers don't go around saying "Oh, that's nigatsu in the year 740, so it may have been more like what I think of now as ichigatsu." The only important thing is to convey to the English reader that the dates are Japanese dates. And if we provide them with a conversion (which I think can done fairly easily), we can actually pinpoint them to the exact date in the calendar they are accustomed to. And by the way, what do you propose we call Leap February for a year where the second month is repeated under your proposed system? -Jefu 23:52, August 18, 2005 (UTC)

Just as an example of how dates can convey different information, the attack of the Forty-seven Ronin took place on December 14 according to the Japanese traditional calendar, and that's barely (or not even) winter. But it was January 30 by the Western calendar, very close to dead center of winter. Similarly, knowing whether a battle or campaign began before or after the harvest can be of interest (if the soldiers also farmed). Fg2 01:58, August 19, 2005 (UTC)

I would support having bracketed info, such as "1973 (Shōwa 48)," but anything beyond that -- as in "January 1467 (February 1467)" -- could get exceedingly confusing. Exploding Boy 00:57, August 19, 2005 (UTC) -- Added: I think that for the majority of the readership, having Western dates be primary in the articles makes the most sense. I'd be quite happy to see a footnote that explained that X Western date correspnds to Y date by the old Japanese calendar, however. Exploding Boy 01:04, August 19, 2005 (UTC)

I don't have a 'proposed system' - I just raised an objection (basically that of the ExplodingBoy addition above). As I presume the unsigned comment saying only "talking only about the months and days" is yours, the worst implication of your proposal is at least nullified, in that the reader still gets a comparative clickable 1473 etc. History articles should really not require the reading of lengthy calender debates though, just to mentally place events 'in the right order'. --zippedmartin 01:16, 19 August 2005 (UTC)

I can't believe we're having this discussion about writiting "day X, 1st month" (say) as "January Xth". There's a very simple reason not to do so: it's incorrect. Day 1 of First Month is not January 1! HELLLLLOOOOOO! This is supposed to be a quality encyclopaedia, not a place where we knowingly include incorrect data because we're too [insert choice phrase] to do it right.

Jefu may have a bit of a point that naive readers may read "3rd day, 4th month" and mentally translate it to "April 3rd". The solution, however, is to stop using the word "month", which gives naive readers the impression that it is the "months" they are familiar with that are being referred to. Use some other word/phrase: "calendar period", "lunar month", whatever.

We have only two viable choices: i) use Japanese dates, and make absolutely clear that these dates are in the Japanese lunar calendar, or ii) convert them properly to the correct Western-style calendar dates. I don't have a strong preference one way or the other; I lean to the second (for the convenience of the bulk of our readers) - but only if we have an easy way to do the correct conversion. Noel (talk) 02:37, 19 August 2005 (UTC)

These slow Wiki servers are really making it hard to post comments. I thought I responded earlier to the question of sites for date conversion, but it was either never accepted or it disappeared. Anyway, the site that I always use for date conversion (and it converts correctly into Gregorian and Julian) is here: [http://www2s.biglobe.ne.jp/~yochicaz/]. The problem is it is in Japanese and I haven't found one in English. Anyway, it is easy to navigate. In the menu on the left there is a section of links that all say "VB Script" or "CGI" after the link. Click the third from the bottom. In the frame that appears top right, the entry fields are year, month, date. The next button is "Convert", the second button is "Clear", the left radio button is to convert Japanese to Gregorian/Julian and the right radio button is to convert Gregorian/Julian to Japanese. The answer appears in the main frame and appears after the "-->". The numbers that result are year, month and date. -Jefu 03:18, August 19, 2005 (UTC)

Interesting. Can you figure out how it works, to make sure it's correct? I.e. does this site really have a complete database of the dai and sho month patterns for each year? (See the link I provided above to my scan of one page of such a table.) If so, is there any way to pull that information out, so we can include it in an article (well, OK, maybe it's a separate page, given how long it's going to be :-) on the Japanese lunar calendar system? Noel (talk) 04:33, 19 August 2005 (UTC)

I just discovered something very cool. If you go back to the left side and click on the last link in the script section (two down from the one I mentioned above) you will get a Japanese to Western only conversion function. This time the input is a box where you input YYYY M D separated by spaces. To input the leap month, you need to input YYYY 閏M D. Anyway, if you put in a nonsense date, the panel below will give you a sort of warning that includes the short and long months in that particular year. So for any particular year you can just enter YYYY 1 31 (which will be a nonsense date for every year in the Japanese calendar). The long months are marked with a 大 and the short are marked with a 小 (the leap month is marked with a 閏 before the month and I'm not sure what the 13 is...). Anyway, I just checked about half of your scanned list and it all appears to be correct. I also sent an e-mail to an e-mail address I found on the site to see if the guy will send me his script. We'll see what happens. I could easily create an English language form on my own site that uses English input and accesses his script, but I'm actually more concerned about the site going bad. He hasn't updated it in several years. -Jefu 05:33, August 19, 2005 (UTC)

And as for Noel's and Fg2's points above, the issue of how to name the months certainly isn't as clear as you both make it out to be. If it is, you're going to have to let all 130 million Japanese, including the thousands of historians here, that they need to relabel their months when speaking of the old calendar to avoid the mass confusion that you are saying would ensue. Japanese have happily been reading for years that the attack of the 47 ronin happened on December 14th (12月14日) even though everyone here draws precisely the same association with that date that anyone who lives in a part of the Northern Hemisphere where there are four distinct seasons. And making people go through the mental gymnastics of calling months first month, second month (or even more ridiculous first lunar month, first calendar unit, first dodecaunit, or whatever), when they're going to mentally translate into the months they are accustomed to anyway, is a very pedantic writing style as opposed to a user friendly writing style, that aids in understanding. The word "February" conveys exactly the same concept as second month, which is the same information that is conveyed to a Japanese reader who reads "２月." The only important thing (and I agree it is very important) is to signal to the reader that the dates are given in a different calendar. Calling it 14th day of the 12th month isn't going to let people know that it was the end of January any more than calling it December 14th will. And if you footnote December 14th and note that this was January 30th by the Julian calendar, this is in no way inferior or more confusing than calling it the 14th day of the 12th month and noting in the footnote that it was January 30th. -Jefu 03:18, August 19, 2005 (UTC)

The thing is that I expect most Japanese know full well that "1st month" means one thing in 1800 and something quite different in 1900. However, I doubt one Western reader in 1,000 knows that. Which is why it's safe to 'overload' (to use the computer science term) the term '1st month' with two different meanings in Japan - and not safe to do so in the West.

And as far as "pedant"ry goes, if that's what it takes to avoid including knowingly incorrect data, well, yeah, we'll have to live with it. Writing "January 3rd" on something when we know that it didn't happen on the day known to most readers as "January 3rd" is just plain wrong.

Besides, if we adopt your suggestion to write "3rd day, first month" as "January 3", then when someone is reading a Japanese-subject article and comes across "January 3", how exactly do you tell if it's really January 3rd (Western), or "3rd day, first month"? People just aren't going to be careful, and always give whatever little mark distinguishes one kind of "January 3rd" from the other. However, if we use a completely different syntax for Japanese dates, and only ever use "January" for Western-calendar dates, there's no problem.

If we want to put in Japanese dates, we have to include some explicit sign that they are in a different calendrical system - "lunar month" is a least a "heads up" that something funny is going on. The other option is to (correctly) convert all dates to Western-system date (Julian or Gregorian, as the case may be, depending on how old they are). Noel (talk) 04:33, 19 August 2005 (UTC)

Let me go home and check my Cambridge History of Japan on this point. I know that they use Japanese dates without conversion, but I don't remember how they translate them. To be honest, I never even imagined this particular issue would be such a sticking point with people, probably because I've been reading Japanese dates that use the same words for both old and new for so long that it never even occurred to me that you wouldn't use January through December when translating into English as well. I mean I can see the point you are trying to make, but it certainly is not so obviously "wrong" to me. I still don't understand what February conveys to an English reader other than "second month of the year." Using anything else just seems far too bizarre for any added benefit. -Jefu 05:33, August 19, 2005 (UTC)

Well, I can't speak for all native English speakers, but to me they have always been the names of months in the Western calendar, with no particular ordering significance. (Ironically, the names "September" - "December" are in fact originally Latin names whose literal meaning is "seventh month" (sic!) through "tenth month" (sic)! In other words, they are off by two!) But I think you've put your finger right on it - when one's native language's name for a month is simply "Xth month", it's easy to take away the meaning as being ordering; but when it's just a random name, with no other meaning, you tend to lose a lot of that sense. Noel (talk) 16:50, 19 August 2005 (UTC)

Okay, it looks like I stand corrected. I stopped at the Maruzen near Tokyo Station this evening and checked out a number of English books on Japanese history. All of them used "Third Day of the Fifth Month" format for dates (one book didn't capitalize the words). The only book that used January, etc. was the Keene book about Emperor Meiji, but they were all converted dates (which makes sense for that period of history, even though the first several years were under the old calendar). It still seems rather bizarre to me, but I guess it is because I'm so used to reading dates in Japanese which uses the same term for modern months and months under the old calendar. Anyway, I'll drop my objection to using numbered months. But I do think the primary dates should be Japanese and the converted dates should be in parentheses or in footnotes (maybe parenthese in text but footnotes when the dates themselves are already in parentheses like the birth and death dates for someone). But I'm still not sure of what to call the leap month. Should it be "Third Day of the Leap Fifth Month" or "Third Day of the Intercalary Fifth Month". I sure don't want to have to use intercalary. Nobody would have any idea what it means. Any other suggestions? -Jefu 11:54, August 19, 2005 (UTC)

There's an article at Intercalation. "Leap month" or "intercalary" could link to it. Fg2 12:16, August 19, 2005 (UTC)

Okay. In that case I guess using intercalary wouldn't be so bad. On further reflection "Intercalary Fifth Month" does sound better than "Leap Fifth Month." although the intercalation article itself needs to be expanded a bit to discuss the insertion of months into lunisolar calendars. -Jefu 12:23, August 19, 2005 (UTC)

Isn't the most important thing to convey the information in such a way that readers will understand it readily? What on earth is wrong with giving the Western date -- we're still talking about the same moment in time after all, and the Western date is the one that will be understood by the majority of readers -- as primary, and providing an explanation of the corresponding Japanese historical date by way of a footnote? Exploding Boy 19:10, August 19, 2005 (UTC)

The biggest reason is that translated dates aren't used by anyone to my knowledge. People reading and studying Japanese history prior to 1873 are going to be exposed to Japanese dates, whether they are reading in English or Japanese. Why should Wikipedia be an exception to that? It is Japanese history after all. -Jefu 00:32, August 20, 2005 (UTC)

Noel as an old hacker I would have thought that you would have suggested date conversion automation. No idea if it works but here is an example:[2]

There are good reasons for putting in conversions, it will help none experts if they wish to compare history in Japan with contemporary history elsewhere. Most English speaking none experts will have no idea when an event took place in Japan if the year is not described in the Gregorian/Julian calender. As to the conversion of the month/day, if it is needed then it should be done correctly (whatever that means see below). For example the campaign season was presumably in the northern hemisphere summer in Japan and, if so, a battle which take place out of the usual campaign season would be be notable. It may well be that experts feel more comfortable with Japanese dates, but those should be included as secondary dates as they are going to be a minority of the potential audience.

In the middle ages in Europe dates are often written as "In the third year of the reign of XXX", (and the day of often given by the saint of the day), but people expect them to be converted into the Julian/Gregorian calender in contemporary articles. (BTW as an aside the battle of Austerlitz is an example of where the simultaneous use of the two calenders had a big impact on history. The Austrians (and the French) were using Gregorian and the Russians Julian so France won the battle as the Russians did not turn up on the expected day!). As by that time the British were using the Gregorian the date of Austerlitz is always recorded in Engish accounts using the Gregorian system. However with dates before 3 September 1752 can be in either and some confusion exists. There is also the problem in Engish of the start of the year, see peypes Diary FAQ "Why do some years appear like 1659/60 instead of just 1660?" the new start of the year was adopted on 1 January 1753. What tends to be the solution for those who bother to worry about it is is that the 1st January will be specified in as the start of year being January 1st, but the day will be mapped as if there was not 12 days diffrence. So an event which took place on 12th of January 1659 old calander, will be mapped to 12th of Jaunuary 1660 in modern texts. This is so that events like the Battle of Agincourt can be thought of as taking place on St Crispins day rather than 12 days later! Philip Baird Shearer 11:52, 20 August 2005 (UTC)

Philip, the problem with automated date conversion support is that our hard-working developers will probably get to it in 2008 or so! Also, nobody (AFAICT) is proposing that we use year dates of the form "Go-Mizunoo 5" - I think everyone agrees that years need to be in BC/AD form, for ease of understanding and comparison.

As to days of the year, in practical terms, for events before 1939, I don't think comparison of days in the year to see if event X in the West happened before or after event Y in Japan is ever an issue, really (with the possible exception of the Russo-Japanese War of ought-whatever - and there you have to deal with the fact that Russia was still on the Julian calendar at the time :-), so I don't think that's a strong reason there. I think Jefu has a very good point that anyone studying Japanese history to any degree at all will soon come up against days in the form "X month, Yth day" - that is absolutely true. I personally don't have a strong feeling either way, as to which of the two day dating systems we should use. Noel (talk) 12:54, 20 August 2005 (UTC)

Philip, I've never suggested that we shouldn't have Western dates. I just don't think we should have Western dates exclusively. And since Japanese dates are used exclusively in the literature (at least in my experience), I think the Japanese date should be the primary date given. We could then either footnote the Western date or put it in a parenthetical. Part of the reason I started this discussion was just to agree on a format for dates that could then be added to the project page. We got sidetracked briefly on wether to use First Month or January, but I think we're back on track now. -Jefu 13:00, August 20, 2005 (UTC)

Anyway, my suggestion is that for dates that are in parentheticals, we put the Western converted date in a footnote. For dates that are in the text we can put the date in parentheses (although the first instance of a Japanese date should be footnoted to explain to the reader that Japan uses a different calendar and provide appropriate links.) -Jefu 13:00, August 20, 2005 (UTC)

And by the way, Japan adopted the Gregorian calendar in 1873 so it is only dates prior to that that are an issue. And other than a few isolated events (Perry's arrival in Uraga is another that comes to mind), I don't think the comparison issue is anything to be concerned about either. -Jefu 13:00, August 20, 2005 (UTC)

Ah, let me make sure I understand your proposal - your reference to "Western dates" and "Japanese dates" made me wonder. As you saying we should put dates in as "1st month, 3rd day, Go-Mizunoo 5" (and then the Western equivalent), or "1st month, 3rd day, 1615" (and then the Western day-of-the-year)?

Also, I see a lot of Japanese articles already seem to use "January", etc when they really mean "First month" - see, for example, Tokugawa Ieyasu.

Finally, in reference our previous conversation about whether "January" has a meaning of "first month" to Westerners - it turns out that before about 1600 or so (it varies from country to country), in the West, the year didn't start on January 1 in most countries! (I had forgotten about this!) Some used March 1, others used other dates! See Julian calendar#Beginning of the year and New Year#Historical dates for the new year for more. Noel (talk) 13:33, 20 August 2005 (UTC)

I think it is very important to use the Western years. Japanese history is divided into too many eras for such dates to make any sense to anyone but the most hardcore historian (even in Japanese, dates are almost always accompanied by Western years to give people some sort of bearing, and Japanese schoolchildren learn important events using Western years.) However, I think we should come up with a format to get in all 3 pieces of information. Since the year is already in the Western date, what if we used a format like this: "First Month, Third Day, Fifth year of Kaei (January 23, 1852)" -Jefu 04:15, August 21, 2005 (UTC)

It seems to me that the main issue here is not so much Japanese vs. English-speaking-world culture, but that the articles in question are part of Wikipedia, which is a huge encyclopedia covering many subjects, all of which are linked together. Because of this, I think it would be a very good idea to use a uniform calendar as the primary one for all articles, viz. the Julian calendar before October, 1582 and the Gregorian calendar afterwards. We can certainly give the dates according to other calendards as a parenthetical addition. Further, Wikipedia is considered to be a work in progress, so there's no reason for dismay in cases where the dates are not yet converted to Julian/Gregorian, especially when this is difficult to do. However, it would probably also be a good idea for us to make it very clear when we are using some other calendar, so as to avoid confusion. I definitely agree that calling the months of the Japanese calendar January, February, etc. is inadvisable. How about just leaving them untranslated? For instance, "November 31, 1582 (3 Ichigatsu, 5 Kaei)"? - Nat Krause 05:22, 21 August 2005 (UTC)

In the English speaking world the change over date is 2/14 September 1752. Even the UNIX command "cal" uses that date! Start of year is confusing but usually kept as January 1st in most texts for all of history. Certainly most computerised programs run using that assumption. One of the propeller-head sports in the year 2000 (Y2K) testing was to check if a program had a year 0 or not. There is also a leap year problem to consider if one does not use the Gregorian clanadar throughout for example was the year 400 a leap year or not? However for day to day usage starting the year on January 1st and mapping to the Julian/Gregorian either side of Sep 1752 is good enough for most of Eurpean mediaeval events if one is not tying in celestial observations. I am not sure what is done by historians for places other than Europe and for European cultures before the high middle ages though. Philip Baird Shearer 15:58, 21 August 2005 (UTC)

I know that for British history the 1752 date is normally the cut-off. But it was always my understanding that continental events between 1582 and 1752 are normally given in the Gregorian calendar. john k 18:28, 21 August 2005 (UTC)

e.g. the Battle of Blenheim is always given as August 13, 1704. Isn't that the Gregorian date? john k 18:29, 21 August 2005 (UTC)

I checked and yes you are correct. It seems that the old joke about New Zealand was true about the UK at this time "You are now disembarking in England please put your watch back 11 days". So William of Orange arrived at Brixham in England on November 5, after a setting sail from the Netherlands on November 11[3]! Irish bashing took place under the "Old Style (OS)" calender on July 1 but is rembered as taking place on July 12 "New Style (NS)". Why William is rembered as landing on November 5 OS but fighting a battle on July 12 NS is odd (I guess in the case of the battle of the Boyne that it is to do with protastants not at first recognising popish dates so they continued to celebrate the aniversery on their protastant 1st of July). Dating and time anomalies still cause some problems, for example the attack on Pearl Harbor and the international date line. It looks from the dates using Zulu time, as if the Japanese waited a day to attack in Asia which is not true, it was on the the same day but a different date! Or the end of the War II in Europe which was on the 9th not the 8th under Zulu time and British double summer time. I have to say that before this discussion I had not given the topic of dates before 1752 much thought, but now my head is starting to heart :-( --Philip Baird Shearer 20:08, 21 August 2005 (UTC)

Wow, what a mess. Personally, I would like to see Wikipedia standardize all dates to Gregorian after October, 1582 while being completely clear about what we're doing so as to minimize confusion. In the meantime, this does present the opportunity for confusion in Japanese history articles that give dates between 1582 and 1752. I still think what we should do for dates before and after is pretty clear. - Nat Krause 09:19, 22 August 2005 (UTC)

I basically agree. Use retrospective Julian and Julian before 1582, Gregorian after. Note old style dates in parentheses where appropriate. john k 16:33, 22 August 2005 (UTC)

Which date goes in the parentheses may seem like a minor point, but I guess the biggest problem I have with using Western dates primarily is the fact that the Japanese date is the only date we can ever really be sure of. In other words, there is no argument among historians that certain incidents in Japanese history happened on certain universally agreed upon dates. But the agreed upon date is the date according to the old Japanese calendar. What those dates convert into isn't always clear as you point out in the discussion above, not to mention the fact that anyone studying Japanese history prior to 1873 is going to come across almost exclusively (and perhaps even exclusively) Japanese dates in the literature. I agree we should include Western dates, but they should just be a guide to the reader to give them bearing using the calendar to which they are accustomed, and to give the author the ability to link those incidents into the rest of Wikipedia. In other words, I think the Western date should be given secondarily and should go in parentheses. -Jefu 01:22, August 24, 2005 (UTC)

Sorry for joining the discussion late, but consensus has apparently been reached on a system I completely agree with. I agree with Noel on almost all points, especially that a Japanese date like Month 2, Day 3 must not be given in an apparently Western form like February 3. For a non-Japanese reader, a date like "February 3" would never be viewed as month 2 day 3. Stating that "February 3" is in the Japanese lunisolar calendar is useless. When Jefu added that note to Emperor Konin of Japan, I did not understand what he meant even though I am quite familar with the traditional Chinese calendar, having studied it for many years. However, the Japanese year should be given in a Japanese style like 5 Kaei or as a number of years since Emperor Jimmu Tenno in 660 BC. Of course it must also be stated in a Western year, but it must be the correct Western year. If the Japanese date is in month 12, the Western year is almost certainly in the year after the year given by a table of nengo years. Repeating the western year found in a nengo table only compounds an existing error for the one or two Japanese months at the end of a Japanese year.

I agree with Jefu that the Japanese date should be given first, and with the equivalent Western date in parentheses or as a footnote (but I would not object to the inverse). The exact order for the elements of a Japanese date is not my concern, although it seems obvious that it should be in the standard Japanese order, if there is one (year-month-day?). The order of the Western date is automatically determined by user preferences when both the month-day and year are wikilinked (Preferences in upper right-hand corner if you are logged in). To use an extreme example, the ISO 8601 form [[1852-01-23]].would be displayed as January 23, 1852 if your user preference is month-day-year.

Although I lean toward "intercalary month 2" (for example); "leap month 2" is also acceptable. "Leap 2nd month" (using the ordinal number) is also OK. The Japanese date could be in Japanese like 12月14日, although an English translation to month 12 day 14 would seem to be preferable in the English language Wikipedia. Note that when using cardinal numbers (12, 14), the units of time (month, day) become adjectives which appear before the numbers.

Some other points mentioned above: Almost all historians use the January 1 to December 31 historical year regardless of the first day of the numbered year (like March 25 or December 25). I do not use "New Year's Day" because throughout medieval Western Europe, New Year's Day always meant January 1, even if the numbered year began on March 25, as in England. King Henry VIII exchanged presents with his court on New Year's Day (January 1) while Anunciation Day or Lady Day, when the year number changed, was just a special religious day. Similarly, Samuel Pepys always called January 1 New Year's Day while changing his year number on March 25. Hence most Western European countries adopted January 1 as the first day of their numbered year while they were still using the Julian calendar. See Gregorian calendar#Beginning of the year. Standard historical practice regarding the Julian/Gregorian changeover date agrees with Wikipedia: That dates before October 15, 1582 be Julian dates and dates after October 4, 1582 be Gregorian dates, unless a particular country adopted the Gregorian calendar at a later date. Then the thorny problem of what calendar should be used in the interim appears.

A standard work used to convert Japanese dates before 1873 into a Western date is Japanese Chronological Tables from 601 to 1872 A.D. by Paul Yachita Tsuchihashi. Whether the online converter given by Jefu gives the same dates, I cannot say.

Although I am quite familiar with the traditional Chinese calendar, including its history and calculation, I am not familiar with the Japanese methods of specifying the date. A big difference between the two forms is that the Chinese were quite concerned that their calendar be in harmony with the sun and moon, whereas the Japanese were satisfied as long as it indicated astrological good and bad days, not caring if their date was two days off. So after the Japanese imported a Chinese calendar in 692, it was used for about a millennium until 1684, while several Chinese reforms were made during that period. See The Lunar Calendar in Japan. In my study of the Chinese calendar, I found the works of Kiyosi Yabuuti very helpful.

— Joe Kress 04:39, August 29, 2005 (UTC)

Footnote, template or something

I and others have been thinking of putting in each article about a Japanese person a footnote or a link near the article title to inform readers about the name order. There was also a talk about dates started by Jefu (good work, Jefu) There are many options to do these things, and I would love to know what people prefer. Here are options I am aware of; please add it if you know one. -- Taku 01:53, August 26, 2005 (UTC)

1. Tokugawa Ieyasu^[1] (徳川 家康; January 31, 1543 (December 26, Tenbun 11) – June 1, 1616 (June 1, Genwa 2)) was a Japanese ruler.
   Junichiro Koizumi[1] (小泉 純一郎; born January 8, 1942) is a Prime Minister of Japan.

   ^[1] is put by {{ref}}. [1] links to a footnote and the footnote talks about name order or date convention, if needed, and links to a longer note. See Template talk:Ref for how to use this footnote method.

2. Tokugawa Ieyasu^[1] (徳川 家康; January 31, 1543 – June 1, 1616) was a Japanese ruler.
   Junichiro Koizumi^[1] (小泉 純一郎; born January 8, 1942) is a Prime Minister of Japan.

   Like the above, but dates in the old calendar are put in a footnote.

Put you comments below:

I prefer (2). -- Taku 01:56, August 26, 2005 (UTC)

I agree, #2 looks good ... I still wonder about which date should be given priority (Western vs. Japanese). I think in an ideal world it would be Western (which is the form the vast majority of users expect), but if the Western date is not always known/calculable (which seems to be the case?) then we might have to go with Japanese. Either way, it probably would be a good idea to explain the Western/Japanese calendar date issue in the Explanatory Note footnote you are working on. My only comment is that the Explanatory "Note" looks like it will inevitably be as long as the Japanese Manual of Style itself, which might kind of defeat the purpose (maybe a link to the J-MoS might work instead?). Or, should each main topic (Japanese names, Japanese dates, Japanese place names, etc.) get its own "explanatory note" for simplicity's sake? Just a few thoughts. CES 02:21, 26 August 2005 (UTC)

I agree we don't want to simply duplicate MoS, but we could make a short page which gives a quick summary of each point, and links to the appropriate reference; e.g. the line about names would just say something like "Japanese names were historically given with the family name first", and link to Japanese names, etc, etc. Noel (talk) 20:47, 30 August 2005 (UTC)

See Wikipedia:Footnote3 for another footnote method, which I think is better than the above. Philip Baird Shearer 21:16, 6 September 2005 (UTC)

n or m?

Wiki standardizes on n because this is what new Hepburn sais. The reasoning to standardize on n.H. from the project page:

This is because it gives the best indication of Japanese pronunciation to the intended audience of English speakers.

Doesn't this strike anyone as a bit contradictory? Current policy creates the necessity to explain for every word with an n that is pronounced as m that it is, in fact, pronounced as m. (Note: I can see the benefits of standardizing on a well-known romanization system, it's just that I think this particular bit is less optimal, especially when considering our target audience.) Shinobu 12:50, 5 September 2005 (UTC)

I absolutely agree. I think that it should be m. Otherwise non-Japanese speakers/students, etc. will look at a word like Tenmu and pronounce it "ten moo" instead of "temmoo". There is no n sound in the word at all. -Jefu 17:14, September 5, 2005 (UTC)

Hmm.. to mee, nm and mm more or less sound alike, especially when said quickly, e.g. Gunma and Gumma. WhisperToMe 18:30, 5 September 2005 (UTC)

I think it should remain as "n." This is how non-native learners learn it, how academic sources romanize, and reflects the actual Japanese spelling. Exploding Boy 17:41, September 6, 2005 (UTC)

None of which has anything to do with the primary goal of romanizing Japanese in this particular context: spelling it in a way such that the non-Japanese speaker/learner can pronounce it properly. The only way "n" makes sense is if the reader understands that the Japanese final n, not followed by a vowel, is not really an n. Nobody without training in the language understands this. That's why, for example, all of the train station names in Japan are romanized with an m, and companies like Asahi Shimbun spell their name the way they do. -Jefu 20:55, September 6, 2005 (UTC)

Regarding the train stations, AFAIK only JR West uses "m" with anything approaching consistency. JR East, JR Shikoku and JR Hokkaido and Tokyo's private operators all seem to use "n". Jpatokal 04:08, 19 September 2005 (UTC)

I really don't think that's true. Think of English words like gunmetal. We don't spell them gummetal, and yet speakers of English can figure out how to pronounce them without much difficulty. Besides which, this is an encyclopaedia, and we should err in favour of being academic. Exploding Boy 21:05, September 6, 2005 (UTC)

I pronounce gunmetal with a distinct n before the m, don't you? -Jefu 07:57, September 8, 2005 (UTC)

I agree with Exploding Boy etc. The phonological difference is too small to make this a worthwhile exception - you'd be better off arguing that transcriptions should drop the 'u' when there's no /u/ realised. In articles where pronunciation is particularly relevant, it should be given in IPA, not a loose transription system like Hepburn. --zippedmartin 22:06, 6 September 2005 (UTC)

I agree with Exploding Boy and zippedmartin. Readers who can't figure out how to pronounce something can check out the Japanese phonology article. And the main focus of reading any article on Japanese culture is most likely not pronounciation, so I agree with Exploding Boy about keeping this encyclopedic. People are likely to come across the 'n' romanization in other scholarly sources, so why not wikipedia? -Parallel or Together? 04:57, 8 September 2005 (UTC)

I don't expect people to achieve perfect pronunciation who aren't students of the language. But I guess my point is that only students of the language would realize that tenmu is pronounced tem-moo and not ten-moo. The romanization of any language, for the typical reader, is for the purpose of representing the language in a form that makes it most accessible to them. This is why Hepburn is so much better than Kunrei for the typical reader. In my opinion, spelling the emperor's name Tenmu rather than Temmu makes it less accessible to the average reader. This is precisely why proper names (which need to be understood the most widely, as opposed to ordinary nouns, etc. which are used almost exclusively by specialists) most often use the m romanization rule (Homma, Asahi Shimbun, virtually all train stations in Japan written for travelers, etc.) I agree that romanizing Japanese with an n rather than an m is vastly superior for language textbooks. But I think you guys are a little bit blinded by your superior knowledge and familiarity with the language and are forgetting what it is like to be a typical reader unfamiliar with the language. And I don't buy the academic argument at all. Perhaps the best and most comprehensive (and scholarly) history of Japan in the English language is The Cambridge History of Japan. And that book uses Emperor Temmu rather than Emperor Tenmu. Although it appears to be a rule in that book only for proper names, since it also uses Tenpyō for the era. -Jefu 07:57, September 8, 2005 (UTC)

I must disagree - I think the academic argument is valid. For example, the Library of Congress uses modified Hepburn (see also Library of Congress - Japanese Romanization(pdf)). Also, the manual of style already includes special exceptions for words like Asahi Shimbun and Emperor Temmu where convention or official policy continues to use the old Hepburn system, a policy I agree with. And I do appreciate your argument that the new Hepburn may be confusing for some people who aren't as familiar with either Japanese pronunciation or the various systems of romanization present. I admit that everytime a Japanese person uses Kunrei (as many of them do) I cringe a little. And it isn't as though Hepburn is without its problems (macrons frequently don't display properly and don't distinguish between おお and おう). Nevertheless, in order to be encyclopedic, I think wikipedia should maintain the modified Hepburn system, altering it slightly for the afore-mentioned cases. If romanizing it as n is superior for textbooks, why not for an encyclopedia? (Also, please don't mistake me for having superior knowledge of Japanese - I am still a beginner and studying it creates new ways of angering me daily). -Parallel or Together? 09:35, 8 September 2005 (UTC)

Some observations:

• We don't use Kunrei-shiki.
• We use Hebon-shiki.
• The MOS (J-r. a.) states "this is because it gives the best indication of Japanese pronunciation".
• We cannot reflect the actual Japanese spelling with roman characters. This is because most words are (completely or in part) spelt with kanji.
• We don't need to reflect the actual Japanese spelling, because usually the real Japanese spelling is shown in the article.

Some thoughts:

• If we want to stick with modified Hepburn, this is not because we want to give an as close approximation as possible to the Japanese pronunciation, but because of the "academic-ness" of modified Hepburn. The MOS (J-r. a.) should reflect this. (I personally don't think a given paper would be considered any less academic if it would spell "sempai" instead of "senpai", by the way.)
• If we want to stick as closely as possible to the kana spelling, we would use Kunrei. I personally don't favour this.
• If we want to stick as closely as possible to the Japanese pronunciation, we would use m. I think slashing u's is a different debate however, and for the moment I'm only concerned with the n or m question.
• I don't like exceptions. However, they may be justified if organizations or persons have decided that their names should be romanized in a certain way.

My conclusions:

• Some change to either the MOS (J-r. a.) or our romanization is needed.
• Given there are good arguments both for m and n, both romanizations would be fine with me (provided the MOS (J-r. a.) doesn't contradict this romanization).

Yours sincerely, Shinobu 15:49, 8 September 2005 (UTC)

I concur, in part, with Gerbrant. Keep spelling things with m when organizations or persons have decided that they want it done that way, but for everything else use n. The n seems like the moderate position, and I really don't want to get into the debate of erasing unvoiced us (and is). That would be ridiculous. -Parallel or Together? 03:43, 9 September 2005 (UTC)

I agree. We don't write です as "dess," just as we don't spell it filay minion. We're not writing a pronunciation guide. Exploding Boy 03:54, September 9, 2005 (UTC)

If we keep the current romanization style, a rephrasing of the following is needed:

This is because it gives the best indication of Japanese pronunciation to the intended audience of English speakers.

How about:

This is because it is generally accepted by scholars and it gives a fair indication of Japanese pronunciation to the intended audience of English speakers.

That would solve the contradiction, I think.

@I really don't want to get into the debate of erasing unvoiced us (and is).: Same here. Although you're wrong to call it "ridiculous". I prefer to leave them in (if only because sometimes they are not so silent), but trying to write down what you hear has been one of the basic principles behind the Roman script. (A fact that Anglophones might have forgotten, but that's because English spelling sucks.)

Yours sincerely, Shinobu 20:39, 16 September 2005 (UTC)

Standardized temple naming

I noticed at Kyoto#Culture that there is absolutely no standardization as to whether temple names are "Foo-ji" or "Fooji". It looks really unprofessional, they way they are mixed up; I think we should pick one, and move the ones that don't fit that scheme to the correct form. (I much prefer "Foo-ji", myself.) I seem to vaguely recall that this was discussed before, but I forget what the outcome was. Can anyone enlighten me? (Oh, BTW, I don't think we should use "Foo Temple", as the "ji" forms are the most common in the West - although one often see "Foo-ji temple"!) Noel (talk) 03:01, 7 September 2005 (UTC)

Don't recall whether there was an outcome. The matter encompasses Kinkakuji or -ji or Golden Pavilion or The Temple of the Golden Pavilion, Kiyomizudera, Sanzenin, Gokurakubo, Sanjusangendo, as well as all the shrines: Yasukuni Jinja, Meiji Jingu or Shrine (which was once translated on a TV show as "Meiji Jing Shrine" if memory serves), Kasuga Taisha, Tsurugaoka Hachimangu (how many alternatives are there for that? Tsuru-ga-oka, Hachiman-gu, Hachiman Shrine...), Kitano Tenmangu, and I can't think of examples, but there's probably a -sha and a -gu as well. And others that escape me. And after temples and shrines there are palaces and villas and castles that could probably be standardized... Fg2 07:35, September 7, 2005 (UTC)

The discussion is here: click here! ... there was no final decision made, but the concensus seemed to be leaning towards Kinkakuji or Kinkaku-ji rather than adding "Temple" or calling it Kinkaku Temple. We talked mostly about temples though, shrines often had different common usage patterns in English than temples. CES 11:36, 7 September 2005 (UTC)

As the one who started the original discussion, I've come to the opinion that the easiest and most unambiguous form is to just use the Japanese name, no hyphens, no English glosses. This is a little more awkward for shrines then for temples though, and I'd suggest that the English "shrine" be used for anything ending in -jinja and -jingu only. Jpatokal 04:27, 19 September 2005 (UTC)

Let me confirm: You're advocating "Tsurugaokahachimangū" and "Fushimiinaritaisha," right? Or do you prefer spaces? And if we space it, there are still the issues of how many spaces, and capitalization. Fg2 05:03, 19 September 2005 (UTC)

Place and personal names should of course be offset with a space, and as these are proper names all words are capitalized. So Tsurugaoka Hachimangū and Fushimi Inari Taisha.

I think this comes down mostly to English ideas of what a 'word' is: a single syllable like -ji or -gu doesn't qualify, but multisyllable constructions like -taisha or -jingu do. Jpatokal 06:12, 19 September 2005 (UTC)

I'm not a big fan of hyphens, but I think the part of the name that corresponds to "temple" or "shrine" should be set off with hyphens. I'm not wed to that approach but it seems more logical than burying it in with the rest of the name. Especially if we go without spaces. -Jefu 05:33, 19 September 2005 (UTC)

Exactly. I think Jpatokal's point about "English ideas of what a 'word' is" is a key one (although I'm promptly going to interpret it differently :-). That principle was used above to separate names up into words; I think we should also apply an 'English word' principle here. The "ji" is a separate word in English, so let's give people who don't know Japanese (i.e. the vast majority of our readers) a hint that it's a separate entity (at least, in English), and use a "-". This is not exactly novel in English works; I don't have time to go through my entire collection of books on Japanese gardens/architecture (it's a whole bookcase!) to do a survey, but I notice that, e.g. Itoh's magisterial Gardens of Japan uses the "-ji" form.

I would apply the same principle broadly for endings which are separate words in English, e.g. "-bashi" (bridge), "-jō" (castle), etc. Noel (talk) 12:58, 20 September 2005 (UTC)

The hyphen is perfect for a name like Tsurugaoka Hachiman-gū, because it operates the way a hyphen operates in English. It separates two words that aren't quite independent of one another, but that haven't quite been fused into a single word with no space between the elements. "Cross-reference" is an example of the former, "crossword" is an example of the latter. And if you were to opt for a translation of that name, you wouldn't translate it into Tsurugaoka Hachimangū Temple. It would be Tsurugaoka Hachiman Temple. The "gū" should be separated somehow, but I think a space is too much. As for whether more clearly delineated words like "jinja" or "jingū" are separated with a space or a hyphen, either would be fine with me, although I would probably lean more toward a space. -Jefu 16:13, 20 September 2005 (UTC)

The October 4 article Hasedera Temple underscores the variety of styles we use.