Talk:Japanese language/Archive 1

From Wikipedia, the free encyclopedia

Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents

Link spam

I am removing all the inappropriate links from here. I recently did the same thing on the kanji page, and put them on a new page learning kanji. Some people objected to this new page as it was inappropriate for Wikipedia. I agree with the objection to the "learning kanji" page, and in the light of that discussion here I have just removed all the spam and irrelevant links. --DannyWilde 00:12, 5 September 2005 (UTC)

Suggestions

I am removing the "arigato comes from Portuguese obligato" line. This has been debunked a million times. I am a near-native Japanese speaker who has lived in the country over eight years.--The Fay.

The statement "In Japanese, a stressed syllable is merely pronounced at a higher pitch" is wrong. Japanese pitch accents are manifested as steep *drops* in pitch. Someone needs to explain the Japanese pitch accent system. Maybe I'll do it myself one of these days when I can find the time...

I second this suggestion. Japanese's so-called pitch accent is one of the most interesting topics is Japanese phonology. I think it would be cool if someone would write a little about this. The Tokyo & Osaka varieties are nicely summarized in
  • Haraguchi, Shosuke. (1999). Accent. In N. Tsujimura (Ed.), The handbook of Japanese linguistics, (Chap. 1, p. 1-30). Malden, MA: Blackwell Publishers. ISBN 0-631-20504-7
An online verison of this is available here: http://www.blackwellpublishing.com/content/BPL_Images/Content_store/Sample_chapter/0631234942%5C001.pdf - cheers! Ish ishwar 05:00, 2004 Dec 22 (UTC)

I think it would be a good idea to include Japanese phonemes here - although of c. not all scholars agree on whether say /S/ (as in Japanese shakuhachi) is actually a phoneme...but the number of phonemes etc. is controversial in most languages... http://pub3.ezboard.com/fhumanjapanesejapanesegrammar.showMessage?topicID=509.topic Wathiik----

If in doubt, go by the writing system. Sure, /S/ will doubtlessly be a phoneme in its own right someday, but for now let's treat it as an allophone. Ditto with all the other palatals, as well as affricates, etc. ..... Besides, it makes the chart that much simpler.
How do you reconcile the difference between sou/shou with your claim that /S/ is merely an allophone? Surely that counts as a minimal pair for the purposes of distinguishing phonemes?
While we're at it, why not change that Assimilation section to a group of footnotes beneath the phoneme table? That way it's a little harder to miss. Ozy 00:42, 2004 Aug 10 (UTC)

I saw your posting and I tend agree with the other guy, Shibatani. So you see there are problems already. I assume he posited the /Q/ because these "double consonants" are proceeded by a very brief glottal stop. Or, maybe he is stuck to the writing system, where the double consonants are written tsu+C? However, if you want to post your analysis, I would link it to this page, which is really not a linguistic analysis.

The analysis with /Q/ opens a huge can of worms, not the least of which are the excessive influence of orthography and the lack of such a "phoneme" in other languages. Also, while some concurrent glottalization is found for the stops (i.e., /p/ /t/ /k/), none is found for fricatives like /s/.
Actually, I tend to agree with the /Q/, even though I don't generally like archiphonemes. /Q/ simplifies analysis of certain constructions, such as 一本 ippon and 一階 ikkai; in both of these, the first morpheme can be analyzed as /iQ/. The alternative analysis -- saying that 一 can be /ik/ only before /k/, /ip/ only before /p/... gets messy very fast. --Marnen Laibow-Koser (talk) 14:40, 14 Feb 2005 (UTC)
Japanese allows only homorganic consonant clusters. There! I've solved the "problem" of the distribution of stops without resorting to the abstraction represented by /Q/. This hardly makes Japanese unusual--just look at Italian for another example of such a language. IMHO, an analysis with /Q/ is much messier than one without. Squidley 16:34, 14 November 2005 (UTC)

[I thought that some info on the history of the language and writing system would be useful, but didn't know where it fit. Other things I would add - A quick note on 'small tsu' stopping, a intro to counters, and a mention of additional blending options in Katakana (vowels and also the newer 'v blends'). Just some ideas.]

There's a brief history of the writing system in the Early writing system section with more at Kana#History of kana. The function of "small tsu" is described in the Hiragana article. Counters are described at Japanese counter word. Katakana are described at katakana. Gdr 22:12, 2004 Nov 18 (UTC)

Are there similar words with Turkish? Like "teppen" and "tepe"? Aknxy 15:55, Apr 6, 2005 (UTC)

What do you mean, Aknxy? Similar to what? I know of no Japanese cognates for "teppen" or "tepe", but I'm not sure if that's what you're asking...? --- Eirikr 06:30, 7 Apr 2005 (UTC)

Sorry. Had to be more precise. I mean words having same or similar meaning and same or similar pronounciation. "Teppen" seems to be one of them. According to dictionary it means: top, summit in Japanese. "Tepe" on the other hand has exactly the same meaning in Turkish. My Japanese teacher was saying that there are some 300 or so similar words like that. I don't know any other examples. But my Japanese is not good. Is this information true? Is it possible? Aknxy 21:55, Apr 21, 2005 (UTC)

Now it's my turn to apologize for how long it's taken me to reply. For teppen at least, the origins of the Japanese word come not from any posited ancient Ural-Altaic roots, but rather from Chinese, as it is a kanji compound like many other Japanese words, and is written 天辺, quite literally "heaven vicinity". So unless there's some Turkish–Chinese relationship happening here, I think you can safely strike teppen off any list of non-coincidental Japanese-Turkish cognates. (Incidentally, 天辺 in Chinese is written either 天邊 in traditional or 天边 in simplified characters, and is written as "tiān biān" using the Pinyin system.) Cheers, Eiríkr Útlendi 19:28, 6 December 2005 (UTC)

Classification

Removed the references to Estonian and Finnish from the text, because it may be confused such that those are Altaic languages. The "Ural-Altaic family" is merely a suggestion; even though Uralic and Altaic languages have influenced each other, the consensus is that no common origin has been demonstrated. --Vuo 02:48, 8 Dec 2004 (UTC)

Japanese is usually classified as an Altaic language. I think there is some scholarly debate about classification of languages re Japanese, but it's usually found under "Altaic" in most family trees. I'm not sure how to effect a change of this? Will find out

Japanese has no proven relation to Altaic languages. Some scholars think it might have such relation but general concensus is that nothing has been proven so far. --User:Taw

The Japanese language is very different from English and most other European languages. Like Finnish, Turkish, and Korean, Japanese is an agglutinative language, with two (phonologically distinctive) tones like Serbian/Croatian and Swedish. It is a language where sentences need no subject and adjectives can have past tenses. It is of uncertain affiliation, though there are theories that it is related to E. Asian languages such as Korean (but not Chinese), though phonological and lexical similarities to Malayo-Polynesian languages have also been noted.

What a awful English-centrism. Sentences need no subject is majority of Indo-European languages. English is weird because they do need subject. Just look at: Latin, Italian, Polish, Russian, any other Slavic language (and many more). --User:Taw

Well, there is a bit of difference there. Those languages all inflect verbs according to subject person/number (ex Latin cogito, cogites), which serves much the same purpose as using a subject pronoun in English (I think, you think). Japanese verbs don't inflect by subject, so if there isn't one explicitly included in the sentence, it's just implied by context. (Of course, that can happen even in English: "Gonna rain" instead of "it's gonna rain", "got milk?" instead of "you got milk?", but these are rarer than subjectless sentences in Japanese.) --Brion VIBBER
Only marking first and second person is obligatory (let's leave issue of number and gender marking). If verb is in third person, the subject is not specified and can be guessed only from context (like in Japanese). --User:Taw
I respectfully disagree; by that logic, sentences using pronouns are sentences without subjects. The use of a third-person form and/or pronoun (which is obligatory in cases where the referred or inferred subject is not first or second person, no?) may require some context to establish the referent, but so do first and second-person: if you don't know who's speaking and who's being spoken to, you don't know who "I" and "you" refer to, do you? A subjectless sentence makes *no* reference, not even a vague one (speaker/addressee/a third party under discussion). --Brion VIBBER, Sunday, July 7, 2002
If we took Japanese, and made a rule that one must use watasi, anata or other appropriate pronoun for 1st and 2nd person, and do as it is now for 3rd person, then we'd end with Polish system. In most situations there is no such thing as third person ending, and you just use base form for that. Even nicer, one can often move 2nd and 3rd person ending to separate word. Example: "widział" (he saw), "widziałem" (I saw), "żem widział" (I saw, colloquial). "gdzie byli" (where have they been), "gdzie byliście" (where have you (plural) been), "gdzieście byli" (where have you (plural) been). As a side note, in some situations you have to use different forms for 1st (kenjogo) and 2nd (sonkeigo) person, but they are also sometimes used with 3rd person. Taw 13:38 Sep 3, 2002 (PDT)
Obviously I'm in no real position to debate with you over Polish. :) But surely you're not going to try to tell me that Latin supports this? "videbam" (I saw), "videbat" (he saw); "ubi erant" (where were they), "ubi eratis" (where were you/pl). --Brion 19:21 Sep 3, 2002 (PDT)

First, "grammatical subject" and "information about subject" are two different things. Most languages don't require first, but usually have some form of second (verb ending, politeness level etc.).

Second, there are sentences which genuinely have no semantic subject, but need grammatical subject in English, like:

  • pada ("it's raining", no "it's" in polish)
  • na stole sa dwa jablka ("there are two apples on the table", no "there" in polish)
  • nie ma cukru ("there isn't any sugar", grammatically it's subjectless possession sentence)
  • nie wolno palic ("smoking is prohibited" , well in English gerund is subject, in Polish it's just normal verb + infinitive, you can't insert any subject here)
  • mozna prosic s¨®l ? ("may I ask for salt ?", it is grammatically completely subjectless again, impossible to insert subject)

There are many other such constructs in Polish, in both colloquial and polite language.

Third, Latin also doesn't need subject, only information about subject. --User:Taw

Apparently, you're trying to say that there is no grammatical subject unless it has its own entire word? --Brion
Grammatical is ambiguous. In Serv-us vid-et there are two markers of subject, -us and -et, but morphologically in entirely different paradigms. One is part of an omissible element, as the sentence Vid-et is also allowed. In the Polish impersonal sentences the verbs are still marked for the non-omissible 'subject' if you want to call it that: sa, = be:3:pl = '(they) are'. In Japanese there is no corresponding non-omissible marker, so it is somewhat different from the common European pro-drop situation. Gritchka
In both Nie ma cukru (there is no sugar - singular) and Nie ma jablek (there are no apples - plural) the same verb type is used and the noun is in accusative. This marker is non-omissible, but it just has to agree with explicit or implied subject or be in right form in subjectless sentence, and no way it is a subject itself. --User:Taw
Which is completely unlike Japanese, in which there is no non-omissible subject marker. What was your objection again? --Brion 16:43 Sep 12, 2002 (UTC)

Check http://pl.wikipedia.com/wiki.cgi?J%EAzyk_japo%F1ski and all related pages on Polish Wikipedia. There is really lot of stuff there. --User:Taw

Hope you don't mind, Tomasz, but I shrunk your "watashi" writing sample from 800x600 to 200x150. (Also, why is "Watasi" in kunrei-shiki but "Romaji" in Hepburn? ;) --Brion VIBBER
In general it's better to recreate image instead of shrinking to get slightly better font quality.

According to sci.lang.japanese faq, Kunrei standard says that Hepburn-written words in Kunrei text are "only to be used for words with strong international connotations, those that are customarily romanised that way or if it strongly improves the information content". I think that "Romaji" and "Kanji" are such words, but don't have strong feelings about that. I prefer kana anyway. --User:Taw

Tomasz, I like kana for two reasons:

1) They are cute.

2) When putting up karaoke lyrics for an international group to sing along with, would you not prefer kana? There are so many romaji systems.

I don't think kana for karaoke would be a good idea. International vocal groups are more likely to be familiar with Latin-based systems that bear some semblance to IPA or SAMPA.

I fixed some formatting in the conversation examples. --Ed Poor

Sino-Japanese Words

This needs to be checked for accuracy before posting:

Sometimes, a word in Japanese both looks and sounds like a word in Chinese. Examples:

Pronunciation (in SAMPA)
Word Meaning Japanese Chinese Korean
愛 love /ai/ /ai/ /{/
存在 existence /sonzai/ /sondzai/Is this right? /dZondZ{/

This is because of borrowing of words from Chinese into Japanese.

However, not all words which were borrowed in this way are alike in today's speech. There may even be false friends. An interesting case is that of the Japanese numeral for ten (/dZu:/), because it sounds like the Chinese numeral for nine (/dZju/ is this right?). (The Japanese numeral for nine is /kju:/.)

ten in Korean /Sip/, nine in Korean between /ku:/ and /gu:/. --Kein Linguist

Examples from Korean, etc., would be good as well. I think they exist.

--User:Juuitchan

I don't speak Chinese (of any variety), but I do have an English-Mandarin dictionary. 存在 is in pinyin cun2zai4, which if the pronunciation guide in the back is to be believed is pronounced something like [ts`u@n/ tsai\]. /sondzai/ might be fitted more easily to the pronunciation in some other Chinese language, of course, which is why you need to specify... Likewise, Mandarin nine is jiu3 [ts`i@u\/]. --Brion
I got it from this: I was listening to a CD with ladies singing in Chinese. I heard a word that sounded nearly identical to Japanese "sonzai". When I looked at the lyrics, I realized that the kanji were the same. --User:Juuitchan
Japanese language borrowed their syllabries and ideograms from Chinese and developed into a system of its own. A kanji may mean something in Chinese, though each kanji refers to different idea in Chinese. No general rule could be drawn for their relationship in meaning. For instance, 私 (Japanese) = I (English) = private (Chinese) that is often not used individually, like 私人 private, 自私 selfish etc. But 大出血 (kanji) means great discount in English and bears no meaning in Chinese. User:kt2
大出血 (kanji) literally means major bleeding in Chinese. That may figuratively describe what is happening to the store owner.  :-) Many Japanese words use archaic Chinese meaning. Most Chinese people may not be aware of those ancient meanings, but people who study ancient literature may tell you that Japanese has preserved many old usages that is no longer known to Chinese nowadays. For example, 優 is an ancient Chinese word for actor. 曜日 were used for weekdays in Chinese but no longer. Kowloonese 00:44, 29 Oct 2004 (UTC)
----
Japanese language has been, like many others, changing over time. I've heard that there are many waves of Chinese influences/ adaptations from China. Most chinese characters used in Japan has two different kinds of pronunciation - on-yomi and kun-yomi, corresponding to pronunciations of Chinese and Japanese origins, respectively. Now there are often multiple on-yomi (Chinese-originated pronunciation) for a single Chinese character, and some of them are attributed to the times those pronunciations are adopted.
The letter "明" (brightness) reads "mei" (as in "文明") or "myo" （as in "光明"）depending on other characters combined with it, and both pronunciations are from china, of possibly different era. The similarity of the contemporary Chinese and contemporary Japanese are, not coincidence. But due to the changes over time occurred for both languages, there isn't neat regularities.
I'm aware of at least three kinds of Chinese-originated pronunciations: 呉音,漢音,唐音.
Disclaimer: I'm a native speaker of Japanese, but not a linguist. Tomos 05:31 Jan 31, 2003 (UTC)
addendum: 存在 pronounces "sonZai," I guess, at least that sounds closer to my not-so-trained ears.
Also, after reading the article and the talk page again, I now come to think that it might help to clarify some more things. Japanese language includes many words imported from China. These days, there is some reverse flow (Chinese importing terms from Japanese - esp. translation of Western words into Chinese characters), as I understand. In other words, at the level of writing, as opposed to pronunciation, there is another reason for the commonalities.
But Japanese has a large amount of its original words. Those are usually related kun-yomi of Chinese characters, the Japanese-originated way of pronunciation. Sorry for this complexity, but what happened was that Japanese people imported Chinese characters, its pronunciations, and many words, while at the same time inventing the way to write their own words using those imported characters. Before importing the Chinese characters, Japanese didn't have writing.
The borrowing has not been only Chinese > Japanese. I have a list, dozens of pages long, of modern Mandarin words borrowed from (post-Meiji period) Japanese, words like "economics" (經濟), "anarchism" (無政府主義). In quite a few cases, Chinese neologisms were later replaced by Japanese ones, leading to homogeneity. Sometimes the Japanese word is a re-interpretation of an older Chinese word to carry a modern meaning (arguably "culture" 文化 is one such word). The new meaning is then exported to China. (The agents of such transfer were probably Chinese scholars studying in Japan.)
btw, Mandarin being a relatively innovative Chinese dialect (and standardized language), it is actually a bad example for showcasing similarities between Japanese and Chinese languages. I.e. Japanese pronunciations are modelled on phonology that is no longer as well preserved in Mandarin. I recommend instead, Taiwanese, Hakka or Cantonese.
A-giau 03:18, 12 Oct 2004 (UTC)
Personally I think Middle Chinese should simply be included -- then readers can compare how Middle Chinese was adopted into Japanese with its eventual evolution into some modern Chinese dialect (probably Mandarin). --Aponar Kestrel (talk) 06:57, 2004 Oct 12 (UTC)

Article Structure

The article looks like still a language guide for English-speakers. While it is not completely wrong, it might mislead people. I don't think for example, Romaji is such a big topic in Japanese language because the Japanese hardly use it in their everyday life. So I will or did already:

  • remove romaji section to romaji article with a pointer.
  • delete duplication
  • add much more examples beyond language learning guide

Or you can help me out to do this if you happen to know something about Japanese language. Cheers! -- Taku 13:32 May 14, 2003 (UTC)

I would disagree that Japanese don't use Romaji in their everyday life, but I agree that the discussion should be moved to another topic page, along with some of the language-guide type content. MattH 08:52 11 Jul 2003 (UTC)

The language is heavily tied to Japanese culture and vice-versa

-- Well, all languages are tied to their respective cultures. Japanese happens to be tied to only one culture (taking as granted that there is only one culture in Japan). Isn't it so? Marco Neves 03:21, 11 Aug 2003 (UTC)Marco Neves

I made a major edit, changing the following:

  1. I added quite a bit of new information (Particularly in the written language section).
  2. I reorganized several sections of the article, in the interest of making the article flow more naturally. The section on Romanization, Unicode and Kanji was moved to the end. The written language is now explained in roughly chronological order.
  3. I removed some duplication, particularly in the written language section, and the sections dealing with kanji, kana, etc.

Things we should still do:

  • Actual Japanese grammar article -- I will start on this soon.
  • Better Japanese language guide article -- working on this.
  • Consider moving Japanese written language to a separate article.

MattH 05:31, 22 Aug 2003 (UTC)

I made another edit. I

  1. redid the tables to make them look neater. I don't think anyone should have problems displaying them.
  2. Added a section on Japanese pronunciation.
  3. Integrated and merged other sections into pronunciation.
  4. made many other small edits.

MattH 07:08, 28 Aug 2003 (UTC)

Writing Direction

I think there's a mistake in this paragraph, but I don't want to change it because I'm not certain:

Up until that time, Japanese text was written top to bottom and right to left. During the Meiji era, the Japanese language first started to be written horizontally. Before World War II, this horizontal text was written from right to left, so as to be consistent with traditional Japanese writing. After the end of World War II, text started to be written from left to right, emulating the common western writing method. Both kinds of writing are still in use today.

I don't think that right-to-left horizontal writing is ever used today. What appears to be right-to-left horizontal writing can be explained as vertical writing in which each column contains only one character. (This theory would be falsified by the existence of a passage written right-to-left which spans more than one line.)

I've never heard of its being used historically either, but that could easily be ignorance on my part.

-- BenRG

The above paragraph seems to say that both horizontal and vertical writing are used today, not both RTL-horizontal and LTR-horizontal. However it may not be worded clearly. --Brion

RTL horizontal writing is still used. Some examples include the engraved characters on some monuments as well as old-style banners and signs. It is rare but occasionally encountered, so deserves a mention.

Brion -- vertical (top to bottom, right to left) and horizontal (just like English) writing styles are both in common use. I tried to edit that section to make it clearer.

MattH

Grammar for Polite Usage

With regard to the section on politeness and "language levels": my impression (I've studied Japanese for nine years but am not fluent) is that there are two independent kinds of politeness level, not just one as the section currently suggests. One kind depends on who you're talking to, and has only two levels, da and desu/masu (which correspond to the tu and vu of European languages). The other depends on who you're talking about (in relation both to you and to the person you're talking to), and it has at least four levels: ordinary, honorific, humble, and contemptuous (yagaru). The two can appear in almost any combination; e.g. for iku the eight possible combos are iku, ikimasu, irassharu, irasshaimasu, mairu, mairimasu, ikiyagaru, and ikiyagarimasu. Of these only the last seems implausible to me, and only because I can't think of any social situation in which it would be appropriate to use it. It's not ruled out grammatically, though.

Does this sound right/wrong to anyone? Should I edit the politeness section to reflect it?

-- BenRG

Teineigo/Keigo is a thorny issue all around, and there are lots of theories on how best to present it to an English speaker. BenRG is right that, as an example, I may be talking informally to a good friend about my teacher, and whenever I mention the teacher I do so respectfully. However, I may also use keigo with the person I am speaking with. Japanese doctors often speak to their patients using keigo. So it does not seem to merely be a question of speech to or about. All the forms may be used with anyone at any time, depending on the circumstances.

I think the best way to present it is with the Japanese concepts. Clearly there are two basic speaking styles: kudaketa (da, verbs ending in -ru, etc.) and teineigo (desu/masu). But this is, of course, something of a simplification, because a good friend may say something like gambarimasu! to his best friend, and the -masu merely adds connotations of seriousness and sincerity. Nonetheless, this dichotomy seems like the best way to present basic formality/informality.

Then there would also need to be a discussion on the particulars of standard, humble, honorific, and even contemptuous speaking styles. Most of this seems to be currently included, but there is always room for more.

While I'm thinking about it, there should probably also be a mention of the differing styles of written Japanese (de aru vs. da vs. nothing at all, etc.).

-- MattH 05:20, 15 Sep 2003 (UTC)

I guess what I'm wondering is: is there any situation in which an honorific verb is used, but the person being honored is not the (explicit or implicit) grammatical subject of the verb?

You mentioned that doctors use keigo when speaking with patients. I would expect this in cases where the patient is also the subject of the sentence, which s/he would be most of the time because people go to doctors to be told things about themselves (or their body parts at any rate).

I found this message (on the Teach Yourself Japanese Message Board) which argues for the to/about idea: http://www.sf.airnet.ne.jp/~ts/japanese/message/message.cgi?sjis;file=jpnDQjvT55HDQjA4nAm.html

The response from TAKASUGI Shinji is also interesting: http://www.sf.airnet.ne.jp/~ts/japanese/message/message.cgi?sjis;file=jpnDQkqngdLDQjvT55H.html

-- BenRG 05:31, 17 Sep 2003 (UTC)

I would say yes to your question. When I ask for o-mizu or o-sake to drink at my friend's house, it is not the subject of the verb. In fact, I may mention any number of nouns in a passing manner with an honorific attached. I would agree that the web page you cite is one way of viewing the politeness situation in Japanese, but I almost always refuse such a neat categorization. Of course, for purposes of presentation, some simplification is necessary. I have edited that section, hopefully for the better.
We also need to do something with this now huge section on the Japanese language and computers. It really deserves its own article. An article on the Japanese language looks lopsided and misbalanced on the Wikipedia with so much information on the computer angle.
MattH 03:55, 25 Sep 2003 (UTC)


Humble Language

"Humble language is used to talk about oneself to one's own group." This sentence is slightly misleading. The purpose of humble language is to elevate the person you are interacting with in the sentence you use it. You aren't really making yourself humble, per se.

Also, what is the difference that is usually made between /wo/ and /o/? Occasionally I hear Japanese people say /wo/ like wo. But most of the time /wo/ sounds exactly like /o/ to my ears.

About the pronunciation I'm really no expert, but I think that Japanese language has lots of subtle variations that happen mainly because in some cases a particular pronunciation is easier to comply phonetically. The fun/pun case when telling time is about the first one a Japanese student will learn, and info it's 100% based on phonetics. Of course there are other cases such as "shyo ga nai" turning into >"sho ga nai", etc.

On the other hand, sometimes there are subtle differences that somehow seem aimed to leave no chance to misunderstandings. That happens often with wo, for example:

"nani wo yatteimasu ka" would almost certainly be pronounced as /o/, while something like "ki ga jukusu nowo matsu" it's pronounced as /wo/. A clear example I can think of is "itta no?" being pronounced as "yutta no?" in order to prevent a misunderstanding between iu and iku.

Someone please correct me, I’m far from being the fittest for this, but I wanted this to be explained here ;)

About ha which is pronounced as wa when used as particle: Yes, that's true but that doesn't mean it should be written as ha in Romaji. In my experience, and I just looked in 3 books and 2 dictionaries that use Romaji, the most common practice is to write it as wa. I propose that our article adopts this convention too. So instead of saying ha and explaining "pronounce it wa", we should write wa and explain "which is actually ha". --130.158.65.240 03:16, 26 Feb 2004 (UTC)

This seems to me to be an excellent article - I'm learning Japanese and there are a number of things which are explained better here than in my textbooks. So thanks to everyone who's worked on it.

I'd be inclined, though, to agree with 130.158.65.240 above, that "ha which is pronounced as wa when used as a particle" seems confusing. Surely the transliteration to romaji means that it should be written (in Romaji) the way that it sounds.

Also, standard practise in the textbooks is to write it (in romaji) as wa. So I've changed the text to "Kochira is the topic of the sentence, indicated by the particle wa (which is written as ha when used as a particle)."

I'm new to the Wikipedia, so if this change is wrong, I'm sure I can rely on someone to change it back. Thanks again for the excellent article.

-- AndyE

This sentence is misleading: "In Japanese, all syllables, with a few exceptions, are pronounced with equal length and loudness." Of course long syllables are pronounced longer than short syllables. I guess the intended meaning is that the length of a syllable is not varied in order to create stress, but that is not what the sentence says at the moment. --Zero 23:38, 4 Apr 2004 (UTC)

A more accurate way to phrase it would be to rewrite that part to use the technical term "mora," instead of syllable.

I added Rendaku to the See Also list. If anyone thinks it's inappropriate, feel free to change it.

-- Nekokaze

Difficulties of Mastering Japanese?

This section seems like myth or nonsense to me. Inter-generational communication impossible in Japanese? 100 core words change every year? 20,000 words needed to read a newspaper? Where do these bizarre ideas come from?

Please provide references or rewrite. User:Gdr 2004 Jul 3

Yeah, I've been eyeing this section suspiciously for a few days now. I think pretty much the entire thing is nonsense. Exploding Boy 00:56, Jul 4, 2004 (UTC)
I second that. I'd be inclined to delete the entire section. That said, there are intrinsic difficulties in learning Japanese, such as the Kanji, levels of politeness and the habit of implying rather than stating things, but then again, all language have their peculiarities. --Auximines 07:43, 5 Jul 2004 (UTC)

But those are not "intrinsic" difficulties. They may not even be difficulties for many learners. Native Chinese and Korean speakers already know kanji. Many languages have varying levels of politeness that rival those found in Japanese. Exploding Boy 07:56, Jul 5, 2004 (UTC)

Well, I would describe kanji as an "intrinsic" difficulty. Chinese and Koreans may not find Kanji difficult, but I've no doubt the rest of the world does :) Anyway, I was trying to agree with you: most of that section is nonsense! --Auximines 09:09, 5 Jul 2004 (UTC)

Yeah, I know. I'm agreeing with you, too. Exploding Boy 15:08, Jul 5, 2004 (UTC)

OK, we didn't get any justification for this section. So I cut it. Gdr 22:17, 2004 Jul 7 (UTC)

I'm assuming that each one of you did researches on your own and felt that it was unjustifiable. I honestly hope that you didn't scrap the entire section in 4 days without any research or notice to any of past writers that you are erasing them. I wrote "20,000 words" but the actual number is 10,000. Check the magazine "Nihongowo Toinaosu" from NHK publishing with ISBN 4-14-189096-0. For the number of new words inducted into Japanese, please read this article about a dictionary published by a company called Sanseido, it's in Japanese and I'll write out important facts.[1]
The 4th edition of this dictionary was published in 1992 and this article is about 5th edition published in 2001, 10 years later. This dictionary has 76,000 words. This dictionary is considered adequate for those in middle school and another one fit for a elementary school student is 33,000 words. In the 5th edition, 3200 new words were added and of those, 200 are onomatopoeia, 1500 newly made words, 1500 previously uncollected words.
In ten years, 1500 new words are added, not including previously uncollected ones. Divide by 10 and you get 150 new words. I wrote "100 new words each year" and I think this is a supportive fact. Does anyone have, other than how one feels or thinks, that each year, Japanese language must have far less than 100 words introduced each year? Remember that most dictionaries doesn't collect words like a name of product, a name of person, or a quickly disappearing slang that can easily double and triple number of new words.
On the inter-generational communication, I probably didn't write clear enough. It should have been something like "Topics that heavily use these new words specific to the certain age group are impossible to communicate without extensive explanations or rewordings with different words." This should not be a surprising fact, imagine telling someone who has no knowledge of recordable CD or DVD to "Why don't you burn the music onto a disc?". Some might say this is becoming an obsolete concept because of i-Pod but lets skip that.
This exists in any language but the situation is far more complex in Japanese language which changes faster. Because of the large stock of words, one can get the meaning across by using very generalized words or by rewording which is somewhat like "simple English" used on some Wikipedia articles. If you are in Japan, in many quiz shows on TV, you can see older guests struggle when a word or a phrase they don't know came up. More than a year ago, there was a quiz section as a part of a popular TV show "Iitomo" that had guests guess what popular words of teenager are and this was discontinued, one of the reason, probably, being that it got so confusing that the viewer rating dropped. Revth 14:44, 13 Jul 2004 (UTC)
I'm sorry, I was too hasty in removing the section. I saw several claims which I know to be wrong or misleading, presented without evidence, and that made me suspicious about the whole passage. I'll put it back and perhaps together we can edit it into something a bit more credible. Gdr 15:19, 2004 Jul 13 (UTC)

Detailed Thoughts

This section considers the claims made and why I think some of them are misleading. Gdr 15:53, 2004 Jul 13 (UTC)

the number of words needed to read and understand over 95% of articles in a newspaper in Japanese is estimated at over 20,000 words while in English, it is only 2,000 words.
More than 2,000 words are needed to understand 95% of articles in an English language newspaper. (I did a quick sanity check on today's Guardian and found 2,810 unique uncapitalized words 24 articles in the UK news section.)
In five articles and two smaller one from the first page of Asahi Shimbun, I counted about 680 unique words not including how to count numbers(I skipped counting words like "29%", "3 people". Former would be counted as 3 words and latter as 2 words in English but they are ignored for a quick counting), including that, actual number is over 700.
Now this is not about how so many words appear on articles. It is about how large "the core of the language" is. If we were to keep counting words that appeared in the entire newspaper and count the days they appeared, English would have highly concentrated "peaks" in such words like "the, a, is, are, have". I'm saying these "peaks" would not be more than 2000 words. In Japanese, "peaks" would be smaller and more evenly spread. These "peaks" are "the core of language" and smaller and evenly these peaks are spread, the language would be harder to master as more knowledge is required. Revth 01:34, 14 Jul 2004 (UTC)
When studying Japanese, the knowledge of Latin that is so helpful in mastering another European language is completely useless.
This is misleading: it doesn't have anything to do with the intrinsic difficulty of Japanese. You might as well say that Japanese is easy to master because it has so much vocabulary of Chinese origin.
But this Wikipedia thing is written in English and not in Chinese in the first place. Also users of Chinese are mostly in China. I also hope you check out Chinese language to realize that while "Chinese characters" have some distinct qualities that are somewhat like what the alphabet is to European language but with many differences, Chinese language is quite different from Japanese language. You may also be surprised to know that Chinese dialects cannot be understand to each other when the distance is too great. Anyway, I don't see why it is misleading to point out that the knowledge of Latin is useless. Take for example a word like "psychology". You know that "psych-" part means something about the mind, "-logy" means "study of" and can deduct that this word must mean "study of mind". In other European languages, the word "psychology" may be spelled differently but they are spelled similar enough to deduce their meaning even if you do not know a word about that particular language. The Japanese word for that is "Shinrigaku", and while Chinese can understand when spelled out in Chinese characters, this is because Japanese translated "psychology" to be "Shinrigaku" and Chinese students who studied in Japan brought them back into China. A simpler example would be greetings like "hello" or "good morning", though I'm not sure if this is from Latin. Check greetings page and you can see that while many European languages are similar, Chinese and Japanese greatly differs. Revth 00:52, 14 Jul 2004 (UTC)
The other obstacle is that the Japanese language changes at an extremely fast pace that would have made it impossible to communicate across generations in any other language. Each year, more than 100 of words are replaced with new words while old words fade away.
This doesn't have anything to do with the difficulty of leaning Japanese. In any case English changes just as fast.
Hmm. I offered the evidence that Japanese changes fast by pointing out that a dictionary got larger by 1500 new words in 10 years. I don't know that if English changes just as fast. Revth 00:52, 14 Jul 2004 (UTC)
Many more new words appear but most of these words will fade away just as quickly. Theoretically in only a hundred years, 50% of words or 10,000 words would be replaced.
This is a big leap. This would only be true if the common words were being replaced at that rate. But in most languages the majority of language churn affects specialist words. Many basic words from Heian era Japanese are the same in modern Japanese: this suggests that the core language isn't changing so fast.
I said "theoretical" and this is just a calculation ignoring that there is a "core of language" that is rarely replaced. I disagree to the second point. Though many basic words from Heian era still exists in the current Japanese, many words have changed their meaning and pronunciation, or I shouldn't have to buy "Kogojiten" lit. old words dictionary, to study for a exam. A novel from Heian era, "Tales of Genji", has two translations in modern Japanese, one being more standard Japanese, other written in "Kogal" Japanese. The latter version was highly praised when it was published for "accurately capturing words used by teenage girls" but they are rarely read now, because words used are so out of touch. Revth 00:52, 14 Jul 2004 (UTC)
The actual pace is believed to be slower, but some have suggested that dramatic improvements in communication technologies may have hastened the appearance of new words. There is even a dictionary that collects words that have disappeared.
This is all true of English too.
A vocabulary of only about 2,000 words may suffice to fully express oneself in English and most European languages (and speak them fluently), however, when it comes to Japanese, knowledge of a mere 2,000 words would be roughly equivalent to pre-school age language skills.
This sentence confuses the notions of getting by in a language and being fluent. 2,000 words is a long way from enough to be fluent in English.
Okay, I missed that point. It should have been "getting by". It should be rewritten. Revth 01:34, 14 Jul 2004 (UTC)

I read with interest this section and the resulting debate, and while there is a kernel of truth to some of the claims made, the whole section is fundamentally problematic because it makes a fundamental assumption that mainstream linguistics rejects; that one language can be objectively and measurably more or less X than some other language, where X is one of difficult, complicated, easy to learn, logical, descriptive etc. It is true that certain aspects of a language can make it easier or harder to learn for speakers of certain other languages, but no language is universally harder or easier than some other language. Certainly Kanji and politeness levels make Japanese harder to learn for English speakers than another Indo-European language, but the claim that core Japanese vocabulary is an order of magnitude larger than English's is frankly laughable.

Secondly, the argument about rate of language change is based on flawed analysis of the data. Saying that edition 2 from year 1991 had X words and edition 3 from year 2001 had X+1000 words; therefore Japanese adds 100 words per year on average is not a reasonable conclusion. How many words that are in a dictionary is much less a function of a size of a language than it is a function of the editorial policies of the dictionary makers. But even disregarding this flawed method of measuring language change, by way of comparison, the Oxford English Dictionary releases a list of 50-100 new English words every quarter. That's 200-400 new words a year, which is larger than your measure of Japanese language change, but as GDR pointed out, the majority of new words in a language aren't core vocabulary but are words at the fringes of language, used only in special areas. Further, the evidence I've read is that the majority of new words in Japanese are English borrowings anyway.

I have more to say about the claims of the size of the 95% core vocabulary, but I have to think more about it.

Nohat 01:51, 14 Jul 2004 (UTC)

A study by the US diplomatic service made three different classes based on the ease of learning language to an English speaker and Japanese was, along with Chinese and Korean, in the third class, most difficult to learn. I don't see why a language cannot be considered more difficult to learn another language.
The point is that no language is universally more difficult to learn than other, which is what the article says (or at least used to say). What I said was "It is true that certain aspects of a language can make it easier or harder to learn for speakers of certain other languages, but no language is universally harder or easier than some other language." Nohat 09:30, 17 Jul 2004 (UTC)
Second, the example I gave was, of a dictionary for a popular usage that doesn't include more technological words or those used in a minor area which the Oxford English Dictionary obviously includes as you point out to yourself. There is a separate dictionary released every few years that collect these words. To negate my claim, you need to provide how many, out of those 200-400 new words, make it into a dictionary, suitably, one that a same age group as the example I gave uses.
Not really. I made the point that this kind of ad hoc analysis is not a worthwhile measure of anything other than editorial policy. You'll have to use some other measure besides dictionary size to demonstrate language change. Nohat 09:30, 17 Jul 2004 (UTC)
Third, I don't see why it is important that majority of new words are burrowed from English. Is this supposed to be, the "X" that means "simpler than"? In Nara era, Japanese adopted many words from Chinese because they were needed. In Meiji era, Chinese adopted many words from Japanese because they were needed. Both China uses Japanese words for their name, translation for "Republic" and "People" were invented in Japan and exported. In Meiji era, Japanese translated many words from French and German for the same reason. In Showa to Heisei era, many English words were adopted for the same reason. They are adopted because they are usable and many of them became "Japanized". To a Japanese, a German word "Karte" (spelled Karute)is used strictly for a sheet of paper used to write down patient's status and not on any other occasion. There is a word "Haripota", shortened term for "Harry Potter" used to describe any of books, movies, or even fans of the series. Also "Pottarian" (same with "Potterian") is in use with the same meaning.
The point of pointing out that most new words are borrowed from English is that for the hypothetical English speaker for whom learning Japanese is more difficult than learning any other language, these new words which have been borrowed from English will be especially easy to learn because he or she already knows them. If they already know the words from their native language, how could they constitute an additional difficulty in learning? Nohat 09:30, 17 Jul 2004 (UTC)

As to the claim,

the number of words needed to read and understand over 95% of articles in a newspaper in Japanese is estimated at over 20,000 words while in English, it is only 2,000 words.

I have to say that the foremost problem with it is that it is unclear what it means exactly. First, I have to assume "read and understand over 95% of articles" means "read and understand over 95% of the words in articles", meaning that having a knowledge of the "core vocabulary" of a language would mean that if you were to read a newspaper, you would recognize and understand 95% of the total words, not unique words. Is this what is meant? It isn't clear. I would like to see a cross-linguistic comparative study that shows that this "95% core vocabulary" is significantly larger for Japanese than for other languages. I would be very surprised to see if that were true. But even if it were true, is this really a measure of the size of the core vocabulary, or is it a measure of the relative difficulty of writing of newspapers between Japanese and other languages. It may be simply a cultural idiosyncrasy that newspapers are written in an especially difficult-to-understand style with baroque vocabulary. In this case, the "95% core vocabulary" measure using newspapers isn't really a measure of the language itself, but a measure of the style used by newspaper article writers.

Nohat 02:06, 14 Jul 2004 (UTC)

I would have to search for a study that says so, but I can present these facts. In a governmental study of Kanji conducted on the year 2000, 8474 different Kanji characters were used in 385 books sampled. Of those, 2457 Kanji accounted for 99% of appearances. 4208 Kanji accounted for 99.9%. In the same study, by knowing only 2602 Kanji, you can read 99.9% of Kanji that appears in the 2 months worth of Yomiuri Shimbun. I think this fact is enough to say that newspaper isn't written in some baroque forms. Also, Kanji are rarely used alone and instead combined with another Kanji, Hiragana, Katakana and even alphabet to form a meaningful word. There are 46 characters of hiragana and katakana plus 26 alphabets, multiply by two because of capitalized letters. So, to merely be able to read the sentence aloud on a newspaper, a Japanese has to know 2746 different characters. Revth 10:09, 15 Jul 2004 (UTC)

Now hold on just a minute. First, let me say, Nohat: excellent post. Regarding this whole business with kanji, 1945 characters are designated for "daily use." These are the kanji a Japanese person must have mastered upon completion of high school. Any written work containing characters not on this list must print them with a phonetic guide. Roman characters are not very often used in newspapers, but in any case that would hardly be an obstacle to English speakers.

Giving estimates of numbers of words is always an iffy business. What qualifies as a word? Some people say the English language has more words than any other (somewhere in the region of 2 million by some counts). But what does that mean in practical terms?

Back to kanji, simply knowing 2000-odd individual kanji is not nearly enough for literacy. Some kanji compounds are intuitive (before and day means the other day, for example), but many are not: "each other" plus "kill" does not mean "to kill each other." It actually has nothing to do with killing at all.

Anyway, I have to agree with Nohat on this one: no language is universally harder or easier than some other language. Exploding Boy 10:29, Jul 15, 2004 (UTC)

I think you are misreading what I wrote. I didn't say that knowing over 2000 kanji let you understand words they form. I wrote "So, to merely be able to read the sentence aloud on a newspaper, Japanese has to know 2746 different characters." To "read aloud" or "pronounce", is one of the two points in what I wrote, other being that newspaper isn't written out in some especially difficult style. To be able to read and understand is another matter, and if we equate being able to read (but not understand) with literacy, Japanese is one of the easiest language to read because hiragana can adequately represent any Japanese words. I didn't touch on literacy part because I needed to look it up(which I haven't done yet).
I wrote about a study that claimed Japanese is a difficult language to learn for an user of English. I believe I originally wrote something like that most natives of European language have a difficulty of learning Japanese because of having have to learn a new and large number of characters unlike learning other European languages where many alphabets are same. Other point is about Latin which I already wrote. I can see I mixed up in the past writing, but that was not my intention and probably carelessness.
While some of things Nohat said is true, he has been selectively using facts when it supported his argument and using indefinite remarks when there is no fact. I gave an example of dictionary recommended for the everyday use of 7 to 9th grade and he comes back to say that OED publishes a larger list of new words. I used this fact to point out that commonly used words changes quite often in Japanese and he comes back to say that the publisher of OED does a good job of collecting new words. Anyway, OED is not a kind of dictionary that a student carry with him to go to school and these two cannot, and should not be compared. Next, I expressed my view that English would have smaller, but extensively used "group of words" (or core words) than Japanese and Japanese would have a larger, but less extensively used group of words. Thus, it would be easier to "get by" with smaller pool of words (I made a correction here). To this, he wrote even if it is true, Japanese newspaper must be written in some baroque vocabulary, an example of a cultural idiosyncrasy and that's why more words would be needed. I'd like to see "a cross-linguistic comparative study" to show that a Japanese newspaper is written in such form. Nohat, please come back with a number from a comparable dictionary and a study that shows Japanese newspaper uses a baroque vocabulary. Revth 18:02, 15 Jul 2004 (UTC)
I will concede that Japanese is a very difficult language for English speakers to learn. In fact, I have long believed that Japanese is probably the most difficult language for English speakers to master. Certainly the memorization of thousands of Kanji is no small task, and neither is the mastery of a grammar that shares no common basis with English. However, I don't believe that the so-called "core vocabulary" is much larger than English, and I have to see any actual evidence that supports that claim.
The main problem is that this question of "core vocabulary" is not well-defined. How large of a corpus are we talking about? What criteria are used for grading the texts in the corpus to make sure they're of the right difficulty? What constitutes a word? What types of tokens are excluded (proper names, abbreviations, etc.)? What other languages were measured and was the same criteria used?
The data you quote is that about 2500 Kanji account for 99% of Kanji encountered in Japanese texts. You also give the estimate of 20,000 or 10,000 words in the core vocabulary of Japanese. Where did these numbers come from? Where did your numbers for sufficient understanding of English only being 2,000 words come from? How are the studies equivalent? Unless you can show a study that compares the two languages in equivalent textual situations, we cannot come to any conclusions about the size of one language's core vocabulary relative to another.
You make the affirmative claim that Japanese core vocabulary is significantly larger than English. If you want the article to say that, you'll have to prove it. It is only my job, in the negative, to cast doubt upon your claims. If you made the claim that there were many more types of mammals than insects and I contested, it wouldn't be expected of me to show a study that shows the number of types of insects is greater than mammals. I could simply say that there is no evidence that shows that there are more mammals than insects. Lack of evidence is sufficient for excluding a claim, at least here on Wikipedia.
Moreover, I'm not making the claim that Japanese newspapers are written in a baroque style—I'm just saying that if it were true, then that might be an alternative explanation for why Japanese is perceived to be so difficult, and we don't have any evidence either way.
As for the rate of language change comparisons, I don't think the rate of language change is significantly higher than any other language and I haven't seen any evidence otherwise. Your ad-hoc analysis of the size of dictionaries is not compelling. But if you insist upon making such comparisons, we can look at the Merriam-Webster Collegiate Dictionary, which is an ordinary dictionary that students might have on their bookshelves. The 10th edition was released in 1998, and the 11th edition, which was released in 2003, claims to have 10,000 new words and meanings, which makes for 1428 new words and meanings per year. If you want to discount the new meanings and only use new words, you can make a very conservative estimate of 10% being new words (and 90% being new meanings), but that's still more than 100 new words per year, about the same as your comparison for Japanese. But like I said, I don't think this kind of analysis is particularly compelling—it is only a measure of dictionary makers' editorial policies and not really a measure of language change. Every language is constantly adding new words. Thousands of words are invented every day. Only some of them catch on, and determining whether a word has "caught on" or not is a function of editorial policy, and does not constitute absolute evidence against which other languages can be compared. Furthermore, even if there were some objective evidence that shows that Japanese has a statistically significant greater rate of change than other languages, that still doesn't constitute evidence that Japanese is harder to learn than other languages. There is no reason to believe that a higher rate of language change makes a language more difficult to master, especially if we're only talking about a hundred words per year or so. That's less than learning one new word every three days; learners of a language probably learn at least an order of magnitude more words than that just while learning the language. I doubt the marginal effect of new words in the language has any impact on the difficulty of mastery.
In sum, I think the difficulty of memorizing Kanji and foreignness of the grammar are sufficient to demonstrate the difficulty of learning Japanese for English speakers without having to make other, more nebulous and less defensible claims about the size of the core vocabulary and rate of language change. These claims should be withdrawn until incontrovertible evidence can be shown that Japanese is indeed more difficult than English in these areas. Nohat 09:24, 17 Jul 2004 (UTC)
Also, I wanted to add that I think the argument could be made that English is just as difficult to learn than Japanese. Even though memorizing thousands of Kanji is difficult, so is memorizing the spellings of English words. Many thousands of English words have spellings that are not predictable from the pronunciation and this serves as a major barrier for learners of English. Furthermore, the stress patterns of English words are not predictable and have to be memorized for almost all multi-syllabic words. When comparing the difficulty of learning English to learning Japanese I don't see any reason why learning thousands of Kanji would be inherently more difficult than learning the spellings of thousands of English words. When you add this to the fact that the phonetic system of English is much more complicated than Japanese, not only in syllable structure, but in the number of consonants and especially vowels, it is quite possible that English is as hard to master or perhaps even harder than Japanese. But please note, I don't have any hard evidence to support this claim, so you won't see me trying to make it in a Wikipedia article proper. Nohat 09:24, 17 Jul 2004 (UTC)
Unscientific evidence this may be, but I've been teaching native Japanese speakers, both adults and kids, English for about four years. In response to the question "which did you find harder, learning kanji or learning to spell in English?" all, without exception, have responded "English." Do with that what you will. Exploding Boy 09:43, Jul 17, 2004 (UTC)

Three-mora Syllables

Are there any examples of these? I was under the impression that Japanese words could have a coda consonant "n" or a long vowel, but not both. To the best of my knowledge this means a syllable can be either one mora ("short") or two ("long"). — Gwalla | Talk 05:44, 17 Jul 2004 (UTC)

Could you give examples for those of us who know Japanese but not linguistics? Exploding Boy 05:59, Jul 17, 2004 (UTC)
Sure. A vowel with or without an initial consonant is a single mora. The "N" is a single mora. The small "tsu" is a single mora (either a syllable-final glottal stop, like the brief halt in the middle of English "uh-oh", or the same sound as the next consonant). A syllable in Japanese can have (in my experience) one or two morae, but not more: てっ, てい, and てん are all valid syllables, but ていん isn't—it'd be interpreted as two syllables ("te-in"). — Gwalla | Talk 03:48, 22 Jul 2004 (UTC)
I don't think it works that way. The terminal "N" you are thinking of its treated as a separate syllable, and you can hear this in music where the singer is stretching a phrase to fit a specific number of beats. Doubled vowels, likewise. The one sticking point is syllables of the form consonant-"y"-vowel. adamrice 22:42, 21 Jul 2004 (UTC)
What some textbooks and teachers call "syllables" are really morae. A mora is a regular beat; a syllable isn't. Listen to Japanese speech—a long vowel is simply long, not audibly two separate vowels. Likewise, there's no syllable break between a vowel and final "N". But syllables with long vowels or final "N" are considered long, with a length of two morae, while open syllables (those without final consonants) with a short vowel are considered short (one mora). I'm pretty sure "doubled consonants" (written with small "tsu") are really (as the article says) a syllable-final glottal stop: I don't think native words can have a doubled consonant after a long vowel or "N", which suggests that the doubling is a final consonant in the preceding syllable and that there is a two-mora upper limit on syllable length. Audible separation of the "N" in singing might be melisma, or it might be a holdover from Japanese poetry (which is based on mora-counting), or it may just be a result of trying to fit lyrics to a clear rhythm. The "consonant+y" is just palatalization of the consonant (the middle of the tongue is raised towards the roof of the mouth so the consonant is released with a "y"-like sound) and doesn't affect length.
So, I think the Intonation section is mistaken: I don't think you can have a three-mora syllable. — Gwalla | Talk 03:48, 22 Jul 2004 (UTC)
Edited the article to reflect this. — Gwalla | Talk 04:12, 27 Jul 2004 (UTC)
Japanese has lots of three-mora syllables. The first syllables in the native words hait.ta and toot.ta; the whole word tooi; Sino-Japanese words like kooi; and loan words like toon, daun, and so forth, all have three-mora syllables. Squidley 16:59, 20 Apr 2005 (UTC)

Honorific Suffixes List

I think a direct link to a list of honorific suffixes in the Politeness section would be nice. I don't know if such a list exists on WP. Although I found many mentions of these suffixes on different pages, I couldn't find an exhaustive list anywhere. I'd really like to see such a list, or at least one with the most used ones and their meaning/usage. --[[User:Gedeon|Ged (talk) (email)]] 23:23, 19 Jul 2004 (UTC)

So far, the most complete list I've found is on the Honorific page. That list is IMHO too long for that page. It should be made a separate page and completed. There is also a (very) short paragraph about these suffixes in the Customs section of the Japanese name page. --[[User:Gedeon|Ged (talk) (email)]] 23:34, 19 Jul 2004 (UTC)


Buddhist Monks Create Hiragana?

I've always been told that hiragana was created by women, based on a simplified version of Kanji. A quick search online elsewhere returned this information: (Taken from www.takase.com) During the end of the Nara period and during the Heian period, literary women (who were not allowed access to the male dominated Chinese learning) developed a syllabary that encompassed all 51 sounds of the Japanese language. This syllabary was based on the Sousho form of the Chinese characters and has a very feminine, flowing form. This style was originally called "onna-de" or "feminine-hand" and is now called "hiragana" and commonly called the cursive style of syllabary.

Obviously, the source is not unimpeachable, but it does mesh well with what I've heard elsewhere. Anyone know a good way to verify this one way or another?

--Beska 16:03, 29 Jul 2004 (UTC)

The hiragana page discusses the history of the script. As I understand it, hiragana was not invented by women, but being the form of writing accessible to them, came to be associated with them. Buddhist monks did create katakana, through a parallel process, by simplifying some of the same kanji used in man’yogana to a constituent element, rather than simplifying to a cursive form. adamrice 16:43, 29 Jul 2004 (UTC)

Historical Kana Usage Page

I've started a page on historical kana usage. Let the flames begin! adamrice


Diphthongs

I'm not a linguist, and I've been taking for granted that Japanese has no diphthongs, however I found out yesterday when trying to explain the "OK" pronunciation to a Japanese that they DO know the concept (二重母音、nijyuuboin. loosely translated, double-weight vowel). As far as I know, Japanese consider things like hyo, byo, pyu, etc, etc. (basically anything that has a small ya/yu/yo) as diphthongs. After some thought, I think it makes sense, but I'm not a linguist. SpiceMan 22:06, 8 Oct 2004 (UTC)

Well, my understanding is that Japanese has two "diphthong-like" phenomena: monomoraic glide-vowel sequences and bimoraic vowel-vowel sequences. The former, which is like your examples of hyo, byo, pyu, etc., are in fact diphthongs from a phonetic point of view, but underlyingly, they're a sequence of a glide followed by a vowel. Compare English yo, which is a glide-vowel sequence, but isn't considered one of the diphthongs of English because the y sound is considered to be a consonant. The same applies to Japanese. The latter type, bimoraic vowel-vowel sequences, are words like hai (="yes"), which again are phonetically diphthongs, but phonologically are considered to be two vowels in separate morae. Japanese has no monomoraic diphthongs, equivalent to English you or cow. I may be off on some of the details here, but I'm pretty sure this is the reasoning behind the "no diphthongs in Japanese" argument. Nohat 20:39, 22 Oct 2004 (UTC)
Imho, "having no monomoraic diphthongs" is not the same as saying "having no diphthongs" at all. I'm going to add something about it. Specially when there are cases which could be called diphthong when considering it in some way (ie phonetically). SpiceMan 09:05, 23 Oct 2004 (UTC)
Nohat has summed up the "no diphthongs" argument nicely. However, the definition of diphthong is basically "two vowels pronounced in one syllable," not one mora. So yes, Japanese does have diphthongs, and like English diphthongs, they are bimoraic and monosyllabic. (Oh, and I am a linguist and professor of Japanese.) Squidley 16:40, 14 November 2005 (UTC)

Pronunciation

I changed some of the nonsensical examples. I don't know how the rest of you pronounce it, but I pronounce "father" with more of an "aw" sound.

I suspect this is more about variations of English pronunciation (US vs UK) than about Japanese....but I think it is wrong to say that the vowel /o/ is like in HOPE. It is more like 90% of HOP and 10% of HOPE. Like these:
http://physics.uwyo.edu/~brent/jal/faq-6.wav
http://www.thejapanesepage.com/mp3/yoroshiku.mp3
http://www.japanese-online.com/language/kana3.wav
--Zero 12:28, 24 Jan 2004 (UTC)

English doesn't have the Japanese vowel /o/ at all. The closest equivalent is what we would say as /o/; namely, SAMPA /@u/. The Japanese vowel is a pure sound, not a glide as the English is. The French /o/ is close to the Japanese one; English is actually quite weird among the world's languages by not having this sound. thefamouseccles
This depends on one's English dialect. The "o" in "hope" in my Northern California dialect is more like /ow/, which is pretty close to pure /o/. /@u/ seems like a British pronunciation. Comparing sounds in any language to English is really vague unless you specify the English dialect, since they vary so much. — Gwalla | Talk 03:58, 8 Feb 2005 (UTC)
Indeed. In fact, I grew up hearing something pretty much identical to the flat Japanese /o/ in the speech of my father, a native of Minnesota and Wisconsin. If any readers of this comment have seen Fargo, you'll have an idea what I mean (though they did carry things to extremes there in the movie). Just ask any Minnesotan to say "Minnesota" and you'll hear it. Eirikr 04:49, 10 Feb 2005 (UTC)
I agree with Gwalla: it does all depend on your accent. Without specifying the dialect comparisons with English vowels are at best vague but at worst can be misleading. Here's what we've got now.
In some English dialects, Japanese vowels can be approximated as follows:
  • /a/ as in father
  • /i/ as in feet
  • /u/ as in soup
  • /e/ as in get
  • /o/ as in go
This isn't all that helpful to anyone, is it? In some English dialects ... in which English dialects? Certainly in Australian English (my dialect) the vowel in go is nothing at all like the Japanese /o/ however the vowel in got is a pretty good approximation. The same would go for New Zealand English.
In AusE and NZE the vowel in go is a diphthong. The same is true in many dialects of British English (including RP) and (as far as I know) South African English too. I wouldn't be surprised to find this vowel to be a diphthong in a few dialects of North American English and Irish English either.
In British Received Pronunciation (RP) the vowel in cot might do but perhaps the vowel in caught would be the closest to the Japanese /o/ but certainly not the vowel in coat. In many North American dialects cot and caught are homophones but where they are distinct the vowel in caught might also work.
Here's what I suggest. Instead of this rather meaningless chart that is supposed to work for some English dialects we have a table showing good approximations in a few different dialects.
Also a note should be made about vowel length. Whilst vowel length might not be phonemic in most English dialects it is still important. Native speakers of most English dialects systematically pronounce the vowels in father, feet and soup as long vowels. The vowels in karate, Nihon and sushi are all short. I wouldn't do just to shove out long vowels in. In fact, you'd probably be better off using the vowels in cut, pit and book.
- Jim 22Mar05

I don't remember where I read this, but I had read that the Japanese vowels /o/ and /e/ are pronounced [ɔ] and [ɛ], respectively, which also matches the vowels as I hear them. Can anyone else verify this?

Example of Glottal Stop?

Can anyone give an example of a Japanese word with a glottal stop? User: Fg2

Yes and no. Glottal stops may accompany geminates (i.e., double consonants) in some words (see below), but are entirely absent in the pronunciation of fricatives like /s/.

  • Gakko (school), kakkoii (cool), kekkon (marriage). Do you need kanji too? Exploding Boy 15:13, Jul 27, 2004 (UTC)
Are those really glottal stops? Aren't they just long k's - /k:/? Tsujigiri 15:22, 27 Jul 2004 (UTC)
Both. What gets called a "long" or "doubled" consonant is, phonologically, a glottal stop (a catch in the throat) followed by another consonant. Phonetically, it's realized either as a glottal stop followed by another consonant or as the second consonant held for longer than it would be by itself, depending on dialect. AIUI, it's considered a separate phoneme because the "doubling" acts like a syllable-final consonant, and phonemes can't span syllable breaks; it's not considered just another /k/ or /t/ because that'd require more complicated rules for syllable-finals (instead of just "a syllable can end in a long vowel, /n/, or /?/", it'd be something like "a syllable can end in a long vowel, /n/, /k/ if the next syllable begins with /k/, /t/ if the next syllable begins with /t/,..." etc.) — Gwalla | Talk 16:05, 27 Jul 2004 (UTC)

Meaning that, since, to take the first example, the word "school" is made up of two words, "gaku" and "ko," when they're put together the sound changes from gakuko to gaKKo (sounds like gak-ko). Exploding Boy 16:31, Jul 27, 2004 (UTC)

There's certainly a stop in gakkō, but is it a glottal stop or a velar stop? Gdr 16:40, 2004 Jul 27 (UTC)

I'm not a linguist, but when I was taking Japanese classes we were told they were glottal stops. Exploding Boy 16:48, Jul 27, 2004 (UTC)

But did your Japanese teacher know the difference? It is hard to hear the velar/glottal distinction as they are both at the back of the mouth, so I doubt that the Japanese make any phonemic distinction. However, I am 95% certain that the stop in ippon is a bilabial stop — it sounds completely wrong if I say it with a glottal stop. If I'm right then the rule "long consonant" = "glottal stop" is bogus. Gdr 17:00, 2004 Jul 27 (UTC)
However, the majority view of the web is that all Japanese geminate stops are glottal. The only dissenting voice I can find right now is [2]. Gdr 17:11, 2004 Jul 27 (UTC)
Not that all Japanese geminate stops are glottal (if they were, kk, pp, etc. would sound exactly the same), but that they consist of a glottal stop followed by a different stop (velar, bilabial, etc.). So ippon would be, phonologically, /iʔpoŋ/ (/iʔ.poŋ/ with the syllable break shown). This is at the phonemic level, though. At the phonetic level they may well be geminate stops: [ippoŋ]. — Gwalla | Talk 18:51, 27 Jul 2004 (UTC)
I understand that phonemically it makes no difference what kind of stop it is. But for the foreigner wanting to know how to pronounce a word, it's the phonetic description that's important. It's at the phonetic level that I am doubting whether native Japanese speakers actually say [iʔpoŋ] and wondering which of [gakkoo] and [gaʔkoo] is right. Gdr 20:19, 2004 Jul 27 (UTC)
This brings up the question of whether it's more important for the description of a language's sounds to be linguistically accurate or to be helpful to language learners. I lean towards the former. It's still possible to figure out how it should be pronounced if those rules are provided (as they are in the current article), and it doesn't gloss over dialect differences. Also, the WikiProject_Language_Template specifies a "phonology" section that discusses major phonological processes, including allophones, which suggests that linguistic accuracy is the goal. — Gwalla | Talk 02:29, 28 Jul 2004 (UTC)

The characters you entered won't display on my computer, but if you're a linguist it should be easy enough to figure it out. What occurs is a doubling of a single consonant: itte (go) is pronounced it.te. Nippon is Nip.pon. Gakkou is gak.ko. Issen is is.sen. Exploding Boy 01:15, Jul 28, 2004 (UTC)

I should've used X-SAMPA: /i?poN/ (/i?.poN/ with the syllable break), phonetically [ippoN]. — Gwalla | Talk 02:29, 28 Jul 2004 (UTC)

Still confused. I've read Glottal stop, and the link from there to Place of articulation, and gakkō sounds to me like a palatal stop: the tongue presses against the roof of the mouth to stop the flow of air. Or velar. I'm not sure where the boundaries within the mouth are. Itte seems alveolar, that is, the tongue presses against ridge behind the teeth. Issen seems not to be a stop, because the breath continues to flow. Ippon is labial, that is, the lips stop the air flow. I cannot think of a stop in Japanese that involves using the larynx to stop the air, as in the English uh-oh. Of course, I'm no linguist, so maybe I've misinterpreted those articles. Like OJ, I'll keep searching for the true glottal stop. Fg2 11:56, Nov 7, 2004 (UTC)

You might be confused about phonemic and phonetic levels of description. Gdr 12:50, 2004 Nov 8 (UTC)

I'm with Fg2 on this. It seems pretty clear from Glottal stop that gakko, itte, ippon, issen etc. are not glottal stops. The true Japanese glottal stop only occurs at the end of some interjections, such as ita' (ouch!) and a' (ah!). These are written with a small tsu (っ or ッ) at the end of the word.

You too might be confused about levels of description. (Hint: is there a glottal stop at the beginning of "apple" in English?) Gdr 13:25, 2004 Nov 8 (UTC)
The interjections like "あっ" do sound like glottal stops to me. As for "apple", I can say it with a glottal stop at the beginning, and I can say it without. Fg2 07:29, Nov 10, 2004 (UTC)
That's right: it makes no difference in English whether you use a glottal stop or not. So although [?ap@l] and [ap@l] are phonetically different, they are the same word to English speakers. So phonemically we can write them both as /ap@l/. I suspect that to a Japanese speaker, [ga?ko:] and [gakko:] are the same word. So it phonemically they can both be written as /ga?ko:/. (However, I do doubt the value of having only the phonemic description of a foreign language in an English-language encyclopedia. English speakers, not knowing the sound system of the foreign language need to see a phonetic description for issues like this one to make sense.) Gdr 12:26, 2004 Nov 10 (UTC)
It's actually explained in the section on phonology (3rd paragraph). — Gwalla | Talk 23:36, 10 Nov 2004 (UTC)

Note about Glottis during Geminate Consonants

I am curious as to what source you are consulting for the phonetic & phonological description of geminate consonants. Can you list the reference(s)? I would like to see something where this was measured instrumentally. Here is a quote that you might find interesting from Akamatsu (1997:156):

"(ii) A warning must be sounded to English speakers to the effect that the first of the two identical voiceless plosives (i.e. [pp], [tt], [kk]) and also the first [c] of [ccɕ] should never be articulated with a glottal plosive, either preceding or simultaneously with the voiceless plosive, still less substituted by a glottal plosive, thus: [ʔp] [ʔ͡p], or [ʔ]. Any use of a glottal plosive together with, or even worse in substitution of, the first of the two identical voiceless plosives or the first [c] of [ccɕ] in Japanese pronunciation is strictly to be guarded against, as it results in an outlandish pronunciation. This warning is necessary because the above-mentioned glottal plosive is widespread in current English, on both sides of the Atlantic, even in R.P., in the pronunciation of [p], [t], [k], and [ʧ] in syllable-final context in English, as in cheap, hot, sack and catch. Those English speakers who use a glottal plosive in one of the various ways indicated above might well carry into Japanese pronunciation their habit of using a glottal plosive in English in contexts where [pp], [tt], [kk] or [tʧ] occurs, as in e.g. cheap potatoes (cf. [kappaʦɯ̥]/[kappaʦɯ]), hot tin (cf. [motto]), sackcloth (cf. [akkoo]) and catch Ann (cf. [ɾaccɕi̥]/[ɾaccɕi]). But this would be as undesirable in Japanese as doing likewise in French in e.g. pompe peinte [põppɛ̃t] 'painted pump', sept tables [sɛttabl] 'seven tables' or avec qui? [avɛkki] 'with whom?', which, if pronounced with the intervention, partial or complete, of a glottal plosive for the first [p], [t] or [k], would be quite unacceptable to French speakers.

Additionally, in footnote #442 Akamatsu (1997:334) states some other previous descriptions by Japanese phoneticians:

"Nomura (1970, p. 703) speaks of 'laryngeal tension or closure' in reference to the first consonants (mentioned just above) in the combinations.... (Norio) Yoshida (1993, p. 19) speaks of 'holding one's breath' (possibly alluding to a closed glottis?) for the duration of a mora while forming an articulation in the oral cavity that is identical with that required for the second consonants. It may be wondered if such references to the said laryngeal state have anything to do with Hattori's view (1960, p. 296) expressed at an earlier date to the effect that the first consonants in question 'correspond to laryngeal tension [my translation]'. Hattori himself (op. cit., p. 753) actually writes in English that /Q/ itself corresponds to the laryngeal tension during the first half of the geminate consonants' or that [ʔ] (glottal plosive) is intended (Hattori, op. cit., p. 416). Note further that both Nomura (ibid.) and Yoshida (ibid.) mention voiceless consonants (e.g. [p], [t], [k]) but not voiced ones. I disagree with a reference to 'laryngeal closure' or 'holding one's breath', as this would correspond to such pronunciations as e.g. [kaʔppaʦɯ̥]/[kaʔppaʦɯ]) (incorrect) instead of [kappaʦɯ̥]/[kappaʦɯ] (correct) for kappatsu 'briskness', or [kaʔssai] (incorrect) instead of [kassai] (correct) for kassai 'applause'. See in this connection my warning given to English-speaking learners of Japanese pronunciation in (ii) above in the text."

References:

  • Akamatsu, Tsutomu. (1997). Japanese phonetics: Theory and practice. LINCOM studies in Asian linguistics 03. München: LINCOM EUROPA.
  • Hattori, Shirō. (1960). Gengogaku no hōhō [Methodology in linguistics]. Tokyo: Iwanami Shoten.
  • Nomura, Masayoshi. (1970). Naihaon [Implosives]. In Kokugogaku Jiten, (p. 703).
  • Yoshida, Norio. (1993). Onsei [Sound] - On'in [Phoneme]. In T. Saeki & Y. Yamaguchi, (Eds.), Kokugo gaisetsu [An outline of the Japanese language], (13th ed.). Osaka: Izumi Shoin. [1st publication date: (1983)].

Anway, this makes me want to investigate this further. Cheers! -- Ish ishwar 00:14, 2005 Jan 6 (UTC)

There is a glottal stop in Japanese(and it is mark orthograpgically with the small "tsu") but it defiently NOT found in the gemmination of voicless stops (p,t,k). The glottal stop found in Japanese is in few interjections like "ah", written 'a' + small 'tsu'. The glottal stop in Japanese isn't a very productive phone, as a matter of fact I don't even believe it is considered a phoneme at all. As far as in the "doubled" stops what simply is happening here is that there is an inaudible release of the stop followed an audible and unaspirated release. quite literally 2 stops. ex: [nip'pon], [gak'ko:]. The syllable structure of those to vords is /CVC CVN/ /CVC CV/ respectively. All this is clearly pointed out in the wikipidea article about the Japanese language. - Alejandro Canizales

Reference

Hi.

I found a study using fiberscope observation where the authors find that the glottis is open during geminate consonant production and furthermore, the glottis is open even wider for geminate consonants than for single consonants (Sawashima & Miyazaki 1973). That is somewhat surprising! So Akamatsu and others are quite correct.

Anyway, interesting stuff. Peace - Ish ishwar 09:02, 2005 Feb 7 (UTC)

For any interested parties, I am providing a quote from Shibatani (1990: 170):
"One might posit the phoneme /ʔ/ for the final sound in interjections such as [aʔ] 'Ah!' and [oʔ] 'Oh!'. When these interjections are followed by the quotative particle to (e.g. [atto], [otto]), the glottal stop also assimilates to the following consonant. Again, in the traditional analysis, these are phonemicized as /aQto/ and /oQto/, but there is no real motivation for this since the glottal stop can be directly mapped on to the appropriate phonetic segment by a well-motivated assimilation rule.
Finally, a brief remark on the phonetics of the moraic consonant. A correlation between the glottal stop and the moraic consonant in the examples above, i.e. [aʔ] [atto], has led a number of phonologists to analyze [aʔ] as /aQ/, and to assume that the basic phonetic value of /Q/ is the closure of the glottis. However, fiberscope observation made by Sawashima and his associates, Sawashima and Miyazaki (1973), indicates that in the articulation of the moraic consonant the glottis is open; in fact, the glottal width for geminates is larger than for a single consonant. Fiberscope observation also reveals that the moraic nasal is different from the mora initial nasals in that the former involves a far greater degree of the velar lowering. This is so despite the fact that the moraic nasal seen in a work like [hentoo] 'answer' in articulatorily quite comparable to the initial [n] in [namida] 'tear' (cf. Ushijima and Sawashima 1972)."
Cheers! - Ish ishwar 15:02, 2005 Feb 23 (UTC)
Another quote about this from Isei-Jaakkola (2004: 24-25):
"In Japanese ‘moraic’ nasals and ‘moraic’ obstruents must be considered separately. With regard to obstruents, it has been suggested that the first part of geminate stops and affricates is usually pronounced with a wide-open glottis (Sawashima & Miyazaki 1973, etc.). The same was confirmed in Finnish stops as well (Iivonen 1975).
Another approach was made in an attempt to gain some insight into the matter of the maintenance of phonation during the articulatory closure of stopped, fricative and affricate geminates in Japanese. A largyngographic (Lx) analysis of the geminates was carried out (Isei 1994), result confirming that the Lx waveform provided no evidence for the postulation of a glottal stop accompanying the first part of the consonant. However, in the waveforms, not only does the amplitude decrease but the valleys between the peaks are also relatively broad, which suggests breathy phonation, unlike impressionistic analysis by some phoneticians or linguists. Thus the results supported Sawashima & Miyazaki’s (1973) direct observation of the glottis."
Biblio:
  • Isei, T. (1994). Electrolaryngographic observation of /Q/. Bulletin Ikuet Junior College (No. 12).
  • Isei-Jaakkola, Toshiko. (2004). Lexical quantity in Japanese and Finnish. Helsinki: University of Helsinki, Department of Phonetics. (Doctoral dissertation, University of Helsinki). ISBN 952-10-1709-0 (pbk); ISBN 952-10-1710-4 (PDF).
  • Sawashima, Masayuki; & Miyazaki, Sachio. (1968). Movement of the larynx in articulation of Japanese consonants. Annual Bulletin (No. 2, pp. 11-20). Tokyo: Research Institute of Logopedics and Phoniatrics, University of Tokyo, Faculty of Medicine.
  • Sawashima, Masayuki; & Miyazaki, Sachio. (1973). Glottal opening for Japanese voiceless consonants. Annual Bulletin (No. 7, pp. 1-9). Tokyo: Research Institute of Logopedics and Phoniatrics, University of Tokyo, Faculty of Medicine.
  • Shibatani, Masayoshi. (1990). The languages of Japan. Cambridge: Cambridge University Press.
- Ish ishwar 16:23, 2005 Feb 23 (UTC)

Weird Edit...

The edit from "218.223.121.169" at 18:22, 3 Aug 2004 is "suspicious" and should (IMO) be partially reverted. It contains at least some legitimate rephrasing (I didn't check them all) but also removes many links and add a broken one. Could someone look into that? I don't have time right now myself... --[[User:Gedeon|Ged (talk)]] 14:08, 7 Aug 2004 (UTC)

Certainly, the wholesale removal of links is suspicious. Fg2 21:26, Aug 7, 2004 (UTC)


Katakana for Chinese and Korean

We had an anonymous edit changing the section on Katakana from: "words and names from foreign languages" to "words and names from foreign languages other than Chinese and Korean".

However, from my experience in Japan, katakana are indeed used for modern loan words borrowed from Mandarin, Cantonese, and Korean, like:

  • ロートル (Japanese: rootoru, Mandarin: 老頭児 [lao3 tou2 er2], meaning 'old')
  • チョンガー (Japanese: chongaa, Korean: 총각 [chonggak], meaning 'bachelor')
  • ワンタン (Japanese: wantan, Cantonese: 雲呑 [won ton])
  • ビビンバ (Japanese: bibimba, Korean: 비빔밥 [bibimbap])
  • ラーメン (Japanese: ramen, Mandarin: 拉麺 [la1 mian4])

Many of these are food words, but I believe it does show that loan words from almost any language, including some from Chinese and Korean, can be written in katakana. Modern loan words almost universally seem to be written using katakana. Of course, a huge portion of the Japanese lexicon comprises direct borrowings from Chinese languages and is written using kanji, but it's not a hard-and-fast rule that katakana is never used to write Mandarin, Cantonese, and Korean loanwords.

--Che Fox 16:39, 27 Oct 2004 (UTC)

I'm going to remove "other than Chinese and Korean" from the Katakana section unless folks think it should stay. --Che Fox 15:43, 28 Oct 2004 (UTC)
Seems ok to remove it, as long as you explain (as you did above) that "a huge portion of the lexicon comprises ... but it's not a hard-and-fast rule ... ." (The words don't have to be the same.) Fg2 22:24, Oct 28, 2004 (UTC)
I agree with Che Fox that the false statement should be removed.
Whether Chinese and Korean is considered foreign or not really depends on the timing of when the foreign terminologies were absorbed into Japanese culture. Words that were adopted centuries ago have already become part of the Japanese language. Many Kanji characters in Japanese have two pronunciations; kun-yomi is Japanese, while on-yomi is the Chinese pronunciation. For instance, the kanji for east, 東, has the on reading tō. However, Japanese already had a word for east, pronounced higashi (or sometimes azuma). Thus, the kanji character 東 had the latter pronunciations grafted onto it as kun'yomi. They had their own word for east, but when they borrowed the Chinese writing for a known concept, they introduced an alternate pronunciation to an old term. Those Chinese terminologies were adopted by taking the Kanji writing and on-yomi pronunciation eons ago and they are no longer foreign. Even these terms should be written in Kanji, if they need to be written in the Japanese alphabets such as in furigana, on-yomi will be written in katakana. However, newer terms like the examples given by Che Fox above are still foreign because they are not part of the Japanese language yet. They are written in katakana despite they can be associated with some Kanji. Whether the Chinese Hanzi or Korean Hanja will be used really depends on how those Kanji characters appeal to the population. In any case, Hangul writing among Japanese text is less likely than Kanji.
This issue also exist in the English language. Words like Taoism, Confucius are considered part of the English language and often written as is. Newer words like Falungong, hanja are still considered foreign and are often written in italic. The use of hiragana and katakana in Japanese language is very similar in nature to the italic use in the English language. Katakana are used for foreign words, period. The exceptions that many Chinese and Korean words do not use katakana are exceptions due to historical reasons, they are not the rules. Kowloonese 00:08, 29 Oct 2004 (UTC)
<Jun-Dai 00:30, 6 Jan 2005 (UTC)> As further evidence, there are a handful of English words that have been sufficiently brought into the Japanese language that they are frequently written in hiragana or even kanji. Consider たばこ, 仙 (せんと), there's a kanji for はいてく (high tech) that I've forgotten, as well as phonetic kanjifications (亜米利加 - アメリカ, 桑港 - San Francisco, 加州 - California, etc.). </Jun-Dai>
Thanks for the comments. I've added a subsubsection Japanese_language#Notes on East Asian Loanwords and Katakana and removed "other than Chinese and Korean" from the Katakana section. Che Fox 04:08, 29 Oct 2004 (UTC)

Language Evolution

How has Japanese changed over the past centuries? That is, what differences are there between modern Japanese and archaic Japanese. What are the present tendencies in the language? 213.226.138.241 13:38, 22 Nov 2004 (UTC)

Well, a common tendency is huge borrowings from english.
<Jun-Dai 00:39, 6 Jan 2005 (UTC)>Among other things, Japanese has undergone some serious revisions to what are officially considered kanji for normal use. Also, many words are no longer generally written in kanji that once were. しまう,　ある (as in ある日), する, ある, いる.</Jun-Dai>
Japanese has undergone quite a few changes, actually. Old Japanese of over a thousand years ago had no long vowels, no /h/ sound, no /ts/, no palatalized consonants, no consonant clusters of any kind, no moraic nasal, nor many of the other things that make Japanese so recognizable. The particles "wa" and "e" were originally "pa" and "pe." There was no "desu" as we know it (!).
Language changes constantly, and there is actually a huge field of research devoted to looking at exactly how languages change and reconstructing what they used to be in order to determine what languages are related to each other. I suggest you look into historical linguistics, if you haven't done so already; we're always looking for more linguists, after all. --Ozy 06:28, 2005 Jan 28 (UTC)
Yeah, right. As someone with a Ph.D. in Linguistics who specializes in Japanese, I try to improve the articles about Japanese on Wikipedia. However, it seems like every time I turn around, whatever I've written has been edited by someone whose knowledge of Japanese appears to have come from a few poorly-remembered classes, or perhaps from living for one year in Japan. Wikipedia is supposed to reflect mainstream viewpoints; whenever I try to balance an article, those who know only one approach remove what they don't recognize (or perhaps it's what they disagree with). It would be nice if Wikipedia lived up to its ideals, which I support. However, the "anyone can edit" sword cuts both ways. Frankly, it isn't worth the effort. If you really want to learn about Japanese, start reading the references listed at the ends of articles. Published works have a lot more authority and accuracy than the web, and with good reason. Otassha de!--Disgruntled Language Fudd
It's a terrible shame if you stop contributing here. Can you point out the problems with the current pages? --DannyWilde 03:23, 15 November 2005 (UTC)

Euro-centric Description of Syllables Misses the Mark.

This description of Japanese syllables (that I found in the "Sounds" section) is overly complex, totally euro-centred, and is likely unrecognizable to any Japanese speaker:

The Japanese sound system is relatively simple, compared to most languages. Syllables generally consist of one consonant (C) and one vowel (V), though many words have longer syllables. Examples include words like kūkō 'airport,' with a CVV.CVV structure (a period indicates a break between syllables); katta 'bought,' with a CVC.CV structure; and tōtta 'passed through,' with a CVVC.CV structure . There are 5 vowel and 17 consonant phonemes (compared to roughly 15 vowels and 22 consonants in English). Japanese syllables consist of:
  • Optionally, an initial consonant, chosen from the 17 consonant phonemes not including ʔ and ɴ,
  • A vowel
  • Optionally, another vowel, which may or may not be identical to the first one, and
  • Optionally, a final consonant. Syllable-final consonants are limited to voiceless obstruents and nasals; only /n/ may appear at the end of a word.

I have moved it here and substituted a description from a Japanese moraic point of view which is obviously simpler (though I'm open to correction). Breaking syllables in the middle of a geminate consonant, while it's obvious to English-speakers, would not make sense at all to a Japanese speaker. It just complicates the language in a way that can't be described with the Kana at all. At the moment I've removed the first description. My real point of doubt is whether the first description has value and should be in the article, or should just be left out.

Steverapaport 14:13, 5 Feb 2005 (UTC)

P.S. If in fact, western linguists insist on describing moraic languages in syllabic terms, then the description above should be included, but prefaced with something like "From a Western point of view, "...

As far as I know, western linguists do describe Japanese in terms of the mora. Perhaps there are also some older or simplified descriptions which overlook this. Sometimes I think people believe that "mora" is a native Japanese word whereas it is in fact a Latin word originally meaning "pause". Does anybody know if there is a specific Japanese word meaning "mora" and not "syllable"? — Hippietrail 21:40, 6 Feb 2005 (UTC)
To answer my own question, "モーラ" is Japanese for "mora", which indicates that the Japanese term is borrowed either straight from Latin or from English. The Japanese term for syllable is "音節". Does anybody know how these concepts were treated in old Japanese linguistics, before being influenced by western linguistics? — Hippietrail 22:03, 6 Feb 2005 (UTC)
OK so it appears the consensus is to leave the moraic description and scratch the other one. I like that. Steverapaport 08:58, 7 Feb 2005 (UTC)
One voice besides your own does not a consensus make! The description of syllables you cut out (and reproduced above) is pretty much standard in the literature in Japanese linguistics, even if different scholars disagree on details. I'd agree that it's not so clear, but perhaps that's because it's an edit of an edit of an edit.... Finally, while breaking up a syllable in the middle of a geminate may or may not "make sense at all to a Japanese speaker," what alternative do you propose? That syllables should start--or end--with geminate consonants? In all the decades I've been reading about Japanese, I've never seen that idea proposed, and I feel confident in saying that no one will propose it anytime soon. So--if you don't break syllables between geminates, where do you break them? Squidley 23:51, 14 November 2005 (UTC)

re: syllable

Hi.

I think that the majority of Japanese linguists of both Japanese and non-Japanese descent consider the mora to be needed in a description of Japanese phonology. A brief look over the literature will show this, I believe (I provided some of references at the bottom of the article). The very first modern linguist that worked on Japanese was Bernard Bloch (1950). He operated with the concept of mora although he used the term "syllable" due to the widespread use of this term in the non-linguistic literature, i.e. traditional European grammar literature (Bloch's "syllable" = "mora"—he mentions this in his phonemics article).

There is debate of whether Japanese needs to be described with syllables or not. Some linguists claim that the notion of syllable is irrelevant to Japanese. Other linguists claim that if the notions of both mora and syllable are used, then a greater degree of generalization can be achieved and it also will account for some exceptions under the mora-only descriptions (this mostly involves explaining the different pitch patterns). Linguists in this camp call Japanese a "mora-counting, syllable language" (McCawley 1978). This issue is still up in the air. So to state briefly, the literature emphasizes the mora and downplays the syllable, which may or may not be needed.

In addition to mora & syllable, other Japanese linguists operate with the notion of foot.

Another thing to note is that a mora could be used as a phonetic unit or a phonological unit or both.

A mora is called haku 拍 in Japanese. I don’t know much about the traditional Japanese analysis. I think that haku is basically something like a mora (but someone should check this out). Even if they are very similar, they both are defined within different theoretical systems. Somewhat off topic: I think I remember reading somewhere that the moraic nasal was not always indicated with a separate symbol (need to check this out, though)—maybe Japanese was not always mora-timed.

There is a very nice summary of this topic in an article by Haruo Kubozono in Tsujimura (1999). Peace. - Ish ishwar 00:41, 2005 Feb 7 (UTC)

If you want to describe the phonotactics succinctly, I think you could list these possible combinations:

  • V
  • CV
  • CjV
  • N   = moraic nasal
  • Q   = moraic obstruent

So, even more compact:

  • (C)(j)V
  • N
  • Q

- Ish ishwar 06:38, 2005 Feb 7 (UTC)

Cool, Ish. what's the 'j' between the C and V? is that like the semivowels 'y' and 'w'? Also I notice that you've added the velar approximant 'ɰ' to the inventory, but a previous contributor mentioned that 'h' becomes 'ç' before i. Is that the same thing or a different one? If different, where does 'ɰ' come in? Steverapaport 08:56, 7 Feb 2005 (UTC)
The wordy illustration of phonotactics that Steverapaport cut and put on this page is basically what Ish has written, except that the wordy version is more concrete (i.e., it doesn't rely on the abstractions that /Q/ and /N/ represent) and is based on the syllable, rather than the mora. Squidley 23:51, 14 November 2005 (UTC)
Hi.
Re: ɰ
/ɰ/ is equivalent to what some write as /w/ and orthographically as <w>. This sound is not phonetically [w] because it does not have lip rounding, so some folks (including me) feel that <ɰ> is a more appropriate symbol. <w> was useful because it was found on typewriters while <ɰ> had to be written in by hand. Computers are getting better at displaying this stuff, so I say use the more accurate symbol.
[ç] is different in that there is audible frication and the constriction is further forward in the mouth. [ç] is an allophone of /h/, but [ɰ] is not.
Actually, [ɰ] is redundant since it is just a short [ɯ] (like [j] is a short [i] and [w] is a short [u]) — we could just use a non-syllabic diacritic.
Re: (j)
(j) is supposed to mean /j/. It doesnt include /ɰ/ — C+/ɰ/+V is not allowed. In other words, Japanese allows an optional /j/ before a vowel and following an optional consonant onset, i.e. V or /j/+V or C+V or C+/j/+V. - Ish ishwar 09:52, 2005 Feb 7 (UTC)
Ok, thanks Ish, I think I got it. Changed the article accordingly -- please check my changes for accuracy and edit them if you wish. Steverapaport 11:37, 7 Feb 2005 (UTC)

re: "Euro-centric" description

So what made it Euro-centric? The fact that the analysis was free of the restraints necessarily imposed by kana? As Ish mentioned, many scholars think that both "mora" and "syllable" are essential to describe Japanese adequately; read the literature and you'll see that. (Quick examples: long vs. short vowels & consonants requires moras, but accent in verbs requires syllables.)

As I read it, the description of syllables that Steverapaport objected to was trying to be all-inclusive, to describe syllables of both single moras and those of more than one mora. That it did not refer to kana is a good thing, IMHO. However, this all seems moot now, as the current article is brief to the point of terseness--and is better for it. Squidley 23:51, 14 November 2005 (UTC)

Pronunciation of "nihongo".

I added a recording of me saying "日本語". Lemme know if it's subpar. - karmosin 07:13, Mar 8, 2005 (UTC)

Can you put it up again in a different file format. While .ogg is technically a clear file format, not that many average users will have a means to play it since it's not normally associated into the standard programs that come with OS's. Could you put it back up in .wav or .mp3 format instead? Ben W Bell 15:28, 8 Mar 2005 (UTC)
I would if there wasn't a decision not to use any other format than .ogg due to issues with licensing. Only formats that can be played in guaranteed free players (if I've grasped the motivation correctly) are allowed at Wikipedia and Commons. Here's the statement from Jimbo. And here are the sound policies.
I know, I know. It's not ideal, but there's not much we can do about it now that it's not even allowed to upload anything but .ogg. Thus spake Jimbo...
What could ease the irritation of unfamiliarity among common users is if someone could edit the audio template so that it could direct the user to a page with clear instructions on how to play .ogg-files. A link from the little speaker image would be a neat way of doing it. I'm not familiar enough with Wikipedia to do it, though. I'm just the guy who can pronounce a lot of stuff... ;-) karmosin 16:50, Mar 8, 2005 (UTC)
Fair enough. Yes .ogg are easy enough to play, but they don't always play straight off the page with most users systems. But the rules are the rules, never mind. Ben W Bell 14:59, 9 Mar 2005 (UTC)
Plugging are problematic, yes, but Winamp can open it lickety-split if you just assign the .ogg-files properly. Works like a charm for me. Almost like playing off the page. - karmosin 15:18, Mar 9, 2005 (UTC)


Adjectives

The article states that "[t]here are three types of words that correspond to adjectives in English: stative verbs (also called i-adjectives), copular nouns (na-adjectives), and the limited set of true adjectives in Japanese." While it's fashionable among linguists to call Japanese i-adjectives stative verbs, I don't believe that this view is at all accurate. About the only verb like behavior they exhibit is inflecting for tense; they still need an actual verb to form a complete predicate (*ringo-wa akai is not a complete sentence).

Um, what? Ringo wa akai is, in fact, a grammatical, complete, acceptable Japanese sentence. It's not natural without a context, but other than that, it's fine. Don't believe me? Ask some Japanese people. (Incidentally, they may prefer ringo ga akai.) If you want an ungrammatical sentence, try *ringo ga akai da. FYI, in the sentence ringo ga akai desu, the desu has no grammatical function whatsoever--it's fulfilling a pragmatic function: it makes the sentence more polite. Compare it with something like, kusaru mae made wa, kono ringo ga akakatta desu 'until it rotted, this apple was red,' where the word akai is inflected; in comparison, *akai deshita is ungrammatical. Squidley 23:23, 19 Apr 2005 (UTC)

The only reason I'm posting this issue on the talk page rather than just changing the article is that, before I edit, I would like to have some clarification on what these mysterious "true adjectives" are -- they're not i-adjectives, and they're certainly not na-adjectives, so I'm really not sure what class of words is being referred to here. --Marnen Laibow-Koser (talk) 14:33, 16 Mar 2005 (UTC)

I would like to have some clarification on what these mysterious "true adjectives" are

Hear, hear. I've been working on Japanese since 1988, and working in Japanese off and on since 1998, and I must admit I'm befuddled as to what "true adjectives" are supposed to be. Anyone more knowledgeable than the two of us care to chime in with an answer?
---
That question aside, I must point out that ringo wa akai is in fact a complete sentence, just in the so-called kudaketa or "plain" form, where verbs take their dictionary endings and the ending copula is da instead of desu. By way of examples for other kudaketa usage, have a look at embedded clauses in any reasonably complex text.
Any textbook that describes how to talk about what you think should also include a good example of kudaketa usage -- the clause leading up to the to particle will usually be in "plain" form. A simple sample sentence would be "I think that's pretty", or sore ha kirei da to omoimasu, with the embedded clause sore ha kirei da, "that's pretty". You cannot grammatically say sore ha kirei to omoimasu nor sore ha kirei, as the na-adjectives require either a following na to connect to nouns or the copula de/da to finish a clause. Now let's look at an i-adjective example, say "I think that's good", or sore ha ii to omoimasu. The embedded clause here is sore ha ii, which you can say on its own as a grammatically complete utterance, for i-adjectives count as a sort of stative verb.
In fact, digging back a bit, it becomes clear why i-adjectives function more fully "on their own" than the na-adjectives. The i at the end actually comes as the modern corruption from the suffix shi, which my copy of Shogakukan's 1988 Kokugo Daijiten states attaches to nouns and various other parts of speech to form the ku-declension (i.e. i-) adjectives (paraphrased translation). You can still see this suffix in modern Japanese in terms like yoshi ashi, or "the good and the bad". So in effect, the i part of i-adjectives is a permanently affixed suffix that plays roughly the same role as the detachable na or da for na-adjectives. --- Eirikr 09:09, 22 Mar 2005 (UTC)
I know that if you dig back far enough, "-i"/"-shi" adjectives start looking like verbs. My comments refer to modern Japanese only. I was under the impression, however, that ringo wa akai, while a common enough colloquialism, was not really considered grammatically correct even in kudaketa, and that the more complete form was ringo wa akai da (rather like the case of "nice day today" vs. "it's a nice day today" in English). Am I wrong? --Marnen Laibow-Koser (talk) 13:48, 22 Mar 2005 (UTC)
Yes, you are wrong. As stated above, ringo wa akai is correct, natural, grammatical Japanese, and *ringo wa akai da is incorrect, unnatural, and ungrammatical.

if you dig back far enough, "-i"/"-shi" adjectives start looking like verbs

Bingo. More on that below.  :)
Regarding ringo wa akai and i-adjectives, I find that people in casual kudaketa speech tend to finish simple i-adjective sentences with the i-adjective (usually followed by some ending particle like yo or ne), whereas na-adjective sentences end in da. Just this morning on the train, in fact, i overheard a few snippets of conversation including "kawaii yo ne!" and "kirei da kedo sa..."
I wonder if what you might have echoing in your memory is the -n da or -no da ending used when making some explanatory comment. The no frequently dissolves into n, which can be easily missed in quick speech if you don't know to listen for it. For example, ii yo ("it's okay") would become ii n da yo, whereas a na-adjective like shizuka ("quiet") would go from shizuka da yo to shizuka na n da yo.
(Incidentally, there is some interesting overlap in how the ending particle yo is used in more archaic speech, and how the ieyo ending copula is used in Korean... but that's a discussion for another time.  :)
To come back to older i-adjectives looking like verbs, you're onto something there. Modern Japanese pedagogy is very visual, very focused on the kanji, to such an extent that folks here in Japan forget that 付く (tsuku, to stick to something, intransitive) and 突く (tsuku, to stick as with a knife or a pole, transitive) are both just tsuku, actually the same word used in different senses. It's as if we had dramatically different spellings for each sense of the word "get" in English, to such an extent that we'd forget that all the "gets" were from the same root.
Part of the upshot of this seems to be a general ignorance for how etymologies can be helpful as a teaching tool. I suppose it's less of an issue with a largely homogeneous population that learns the language from birth, but I find it interesting the contrast between approaches to language learning and education, specifically between what I see here in Japan (both teaching foreigners and teaching children) and what I grew up with in the US. In elementary school, for instance, I was taught to break words down to the roots, and that I could learn new words and learn how words are related by looking at the roots. Take "homogeneous", for instance -- "homo", same, "gen", as in "genus" or kind, "ous", adjectival suffix meaning something like "full of" or "lots of". However, try to do the same thing with Japanese words and ask about the parts with most native Japanese speakers, and they either look confused or thoughtful, as if presented with something new. No dictionaries I've seen include any etymology for the vast bulk of the words listed. And yet the connections are there.
Take akai, for instance. (This is one of the few where I've found some etymology information.) We know that the final i is what makes it an adjective; this leaves us with the stem aka, which means "red", as in the noun for the color. Other meanings attached to this sound include 1) some sort of dirt or grime, possibly a cognate in terms of additional color, and 2) a connection with aka as in 明 (listed as a related root for akai in a number of dictionaries I have consulted), the stem for akarui "bright". In this sense, the word is connected to 開く aku, "to open", from the opening of the day -- dawn. From here, we can jump to ideas of emptiness -- 空く aku, "to become empty" as a twist from "to open up", thence to 飽きる akiru, "become fed up with" as a twist from an older meaning, "to have enough (and thus be free from want)", and so on and so forth.
Anyway, forgive my long tangent, but my basic point was that, if you feel like you're seeing more under the surface, you'd be right.  :) Cheers --- Eirikr 09:31, 23 Mar 2005 (UTC)
I meant to add this a while back... I just wanted to say that using da with an i-adjective may even be considered bad Japanese. In my shodou class someone once said (about the food on a certain restaurant) Takai da ne. Sensei (which is a native speaker from Tokyo, originally from Saitama) almost burst into laughter -- she said that sounded like countryfolk speech (she used a funny, mildly derogatory term in Spanish). Upon questioning she confirmed that we should never ever say Takai da ne, although of course, Sou da ne and Kirei da ne are alright (the problem is with i-adjectives/stative verbs). --Pablo D. Flores 14:00, 6 Apr 2005 (UTC)
Interesting. Did your teacher say that "takai ne" would be the accepted usage, or did she suggest another form? And is "takai da ne" grammatically incorrect or just considered awkward idiom? --Marnen Laibow-Koser (talk) 17:12, 7 Apr 2005 (UTC)
I'm not sure. I was left with the very clear impression that takai da ne was nonstandard and at least uneducated. I don't know if, for a native speaker of a language with considerable dialectal variation, there is a clear line between funny/awkward and plainly ungrammatical. But, from the point of the view of a Japanese teacher trying to get non-natives students to learn good Japanese, i-adjective + da seems to be a definite no-no. --Pablo D. Flores 20:44, 7 Apr 2005 (UTC)
There are Tohoku dialects which have a sentence-final particle da which is probably related to na. In those dialects, they can say things like soo da da, but the important thing to remember is that the first da is the copula and the second is not. The second one fits in the same class as yo, ne, etc. That's why your teacher thought it sounded rustic. For speakers who are not familiar with those Tohoku dialects, it's just wrong, and in any case, it's not part of standard Japanese. -- Japanese teacher

Kansai-ben

Kansai-ben is very close to the standard language, and varies essentially in slang; some consider it to be equivalent to the standard language

Does anyone besides me find this to be a pretty bizarre statement?

  • Kansai-ben (and most dialects of Japanese) vary in many significant aspects (intonation, sentence formation, word choice, etc.)
  • There are many differences besides just slang (maybe the author meant plain/informal speech by "slang"?)
  • Who on earth considers Kansai-ben to be equivalent to the standard language?

I could see a statement like the above applying to, say, the Yokohama dialect, but I'd have a hard time equating Kansai-ben with the "standard language."

CES 16:37, 23 Mar 2005 (UTC)

Yes, it is a weird statement. Please fix. One thing very striking thing to me is that these speakers use a different pitch accent system. The melody of nouns in Osaka is often the opposite of Tokyo speakers.
(Regarding standardness, this is just politics. Who on earth considers any variety to be the standard language? But, this is a topic for sociolinguistics in general...)
peace. — ishwar  (SPEAK) 17:31, 2005 Mar 23 (UTC)
The Japanese government until recently (last decade or so?) considered an elevated form of the Tokyo/Kanto dialect to be the hyōjungo, the Standard (Japanese) Language. I think nowadays they call it futsūgo, Ordinary (Japanese) Language, to be fair to the dialects and to Okinawan and Ainu. Most Japanese educated before around 1985, particularly those in the Kanto region, seem to continue to believe the old political line that dialects are bad and the Official Language is the only decent Japanese, everything else is condemnable. The views change in other areas and with the younger generations. I have a reference on this from a sociolinguistic study a couple years back, but I'll have to hunt for a while to find it.
In any case, Kansai-ben is the same as Kanto-ben only in that it's a Japanese dialect and thus has a similar inventory of N/Adj/V, similar syntax, and similar morphology. The two are quite different when you get down to particulars. See the Kansai-ben article for more info. — Ts'éiyoosh 02:10, 25 Mar 2005 (UTC)

I removed the entire paragraph, it seemed inaccurate and POV. CES 22:31, 25 Mar 2005 (UTC)

Left, Right, Top, Bottom?

I've seen Japanese katakana and kanji written left-to-right on pages, but I've also seen just kanji written top-to-bottom, with the columns right-to-left. I also notice that Japanese books are read back to front. Could information on this be included in the article? --Poiuyt Man (talk) 08:13, 30 Mar 2005 (UTC)

That wouldn't really be within the scope of this article -- this article is about the language, not the script. See Japanese writing system for the script. --Marnen Laibow-Koser (talk) 13:24, 30 Mar 2005 (UTC)


Origins of Standard Japanese?

I was looking for information about which dialects/regions formed the basis for modern Standard Japanese, but the article doesn't appear to cover that. -- Danny Yee 01:16, 22 Apr 2005 (UTC)

Hmm, that should probably be added at some point. The Japanese government of the early Meiji Period did set out on an ambitious project to formulate "standard" Japanese for use in education, based on a survey of the language throughout the country, but, like many such complex and enormous initiatives, this fell by the wayside. What was eventually adopted for educational purposes was the speech of upper-middle class speakers in Tokyo. If you can get your hands on a copy of Masayoshi Shibatani's book The Languages of Japan, he gives a decent summary of what happened. Any good history of Japan of the Meiji Period should also include some mention of this process. Cheers, --- Eiríkr Útlendi 02:27, 22 Apr 2005 (UTC)

Learning Japanese

Is it just me or is the "Learning Japanese" section short on actual information and long on POV? It seems like it needs to be cleaned up or deleted. CES 02:55, 22 Apr 2005 (UTC)

Transcribing japanese to english, not translating

What is the term for transcribing japanese to english, so that you can read it in english, but pronounce it as if you were reading it in japanese? Edit: and are there any online programs to do this? Babelfish just returns japanese characters.

It's called romanization. See romaji for detail. --Kusunose 13:25, 2 May 2005 (UTC)

A request for assistance.

Greetings. Would someone who can speak Japanese be kind enough to have a look over at the Reiki article? There seems to be a dispute over the actual meaning or meanings of the usage 霊気 reiki itself in Japanese. Also, there is no clear provenance as to whether the term is a recent coinage or a term from earlier Japanese history. Any help would be greatly appreciated. Regards, Fire Star 04:34, 3 May 2005 (UTC)

According to my dictionary, 霊 rei (another reading is ryou) means "soul, spirit, departed ghost". It appears in other compounds such as akurei or akuryou "evil spirit", eirei "spirits of war dead", onryou "revengeful ghost", jorei "exorcism" and sourei "poltergeist". So it doesn't mean "universal" as the page states. --Pablo D. Flores 11:21, 3 May 2005 (UTC)

Proposal to move Japanese language#Sounds to Japanese phonology

Would anyone strenuously object to moving the Sounds section to an independent article? The current article is 40kb, and the phonology section is sufficiently detailed to stand well-enough as an article of its own (IMHO). Tomer TALK 01:16, May 27, 2005 (UTC)

I agree. The article should be more than just about grammar or phonology. Spliting can create a room for history, culture and people related to Japanese language. -- Taku 01:20, May 27, 2005 (UTC)
Breaking out the section and leaving a short summary behind would be great. — Gwalla | Talk 04:55, 27 May 2005 (UTC)
Support--Jondel 05:10, 27 May 2005 (UTC)
Support. I've been doing this myself for a few long article. Remember to leave a {{mainarticle|Japanese phonology}} tag. --Pablo D. Flores 11:14, 27 May 2005 (UTC)

OK. I moved it. Anyone got any ideas for a mini-intro to put into the #Sounds section so it doesn't look quite so bare? Tomer TALK 13:05, May 29, 2005 (UTC)

Question for Japanese speakers

Over at Teikyo University it says the university was originally called Teikyo Commercial High School (帝京商業高等学校; teikyo syogyo kōkō). Is that right? I can't read kanji, but it looks to me like there are too many kanji characters for that romaji transliteration. If it's wrong, could someone correct it at that page? Thanks! --Angr/tɔk tə mi 29 June 2005 23:38 (UTC)

It's 帝京 (ていきょう, teikyō) 商業 (しょうぎょう, shōgyō) 高等 (こうとう, kōtō) 学校(がっこう, gakkō), or Capital (city) Commercial Senior High School. Exploding Boy June 30, 2005 01:07 (UTC)


Mistake?

This section of text is from the article.

"desu です is a contraction of de で" should this be changed to "desu です is a contraction of de gozaimasu で ございます"?

Markcox 6 July 2005 04:37 (UTC)

de gozaimasu is used when talking to customers or the emperor, etc. Desu is not a contraction of de gozaimasu.--Jondel 08:55, 14 July 2005 (UTC)
is not the first time i've read/heard that "desu です is a contraction of de gozaimasu で ございます", the first time was when my japanese language teacher told me that. On the other hand asking about the subject to a regular japanese in the fashion "is desu a contraction of degozaimasu?", gets you a "へー、気つかなかった、なるほど！" (wow, I haven't noticed before!), or something similar. I've also read that it's a contraction of であります. Which makes as much sense as the former. Be it the former, the latter, or none of them is something hard for me to grasp, but is something frequently read/heard. Any japanese native which in turn is a japanese language scholar around? SpiceMan 15:47, 14 July 2005 (UTC)
The article currently says, "Although there are many theories regarding the origin of desu, the general view amongst linguists is that desu です is a contraction of degozaimasu でございます." This is a lot better than what it said before (thanks) -- but is it even true that degozaimasu is the "general view amongst linguists"? I can't find this claim outside of the sci.lang.japan FAQ that DannyWilde mentioned below.
The Daijisen dictionary says: 「です」の語源については、「で候」「でおはす」「でございます」「であります」など諸説ある。 ("For the etymology of 'desu', there are various theories: 'desou', 'deohasu', 'degozaimasu', 'dearimasu', etc.") The Daijirin dictionary says pretty much the same thing: 「です」の語源については、「でそう（で候）」説、「でござります」説、「であります」説、その他があるが、まだ定説化されたものはない。 ("For the etymology of 'desu', there is the 'desou' theory, the 'degozaimasu' theory, the 'dearimasu' theory, and others, but there is no generally accepted theory yet.")  --mconst 15:50, 11 August 2005 (UTC)
This is a FAQ on the newsgroup sci.lang.japan: see sci.lang.japan FAQ page I think the answer is that nobody actually knows for sure.--DannyWilde 00:56, 15 July 2005 (UTC)

であります。is polite. でございます。is super polite(keigo).Like imasu and irasshaimasu. If you are a guest in a hotel, and the front desk needs to know your nationality, he would say: Amerika jin de gozaimas ka? If there is a Japanese you've casually met in a restaurant, he would say, Amerika jin des ka? A very young Japanese child hearing you speak English would ask , Amerika jin ka?(rude)

Japanese teachers seem to teach very theoretical/old/classical Japanese. I was taught that the polite form of cheap (yasui) and blue (aoi) was yasoo gozaimsu.(安そうございます。) and aoo gozaimasu(青おうございます。)　But only a few very old Japanese know this. The rest tell me they don't know this. Who is right? To be on the safe side master the polite(textbook) form then learn the conversational Japanese. --Jondel 00:52, 15 July 2005 (UTC)

It's not that they 'don't know' it ... the ~ou gozaimasu form is so rarely used and is not really a useful conjugation anymore. It exists only in a few traditional phrases like ohayou (from hayai) and arigatou, (from arigatai). Regular adjectives are not conjugated in this way. It might be in some way grammatically correct to say that the book is 'aoude gozaimasu', but laughter will result as you're using formal speech that hasn't been in common use for hundreds? of years. It IS good to know about the ~oude gozaimasu conjugation, or rather it may be interesting, if you're interested in the origin of some Japanese expressions, but it's about as useful as knowing that 'goodbye' comes from 'Good be with you' (or whatever the correct expression is).

---

Naruhodo.--Jondel 04:24, 15 July 2005 (UTC)

About the -u form. It is actually the onbin 音便 form of "adverbs" or similar words.

  • arigatoo = arigataku
  • yoo = yoku
  • ureshuu (うれしゅう; before spelling reform うれしう) = ureshiku

the last one of which is due to the fact that the shi + u classical Japanese combination is pronounced shuu. This kind of onbin was prevalent in late-ish classical Japanese, which saw the so-called -te forms of verbs having an additional alternative onbin form in addition to the ones we are familiar with in modern Japanese, e.g. for omoo (思ふ; おもふ), omotte (おもつて) = omoote (おもうて) (standard form omoite おもひて); similarly koo (買ふ; かふ) => katte (かつて) = koote (かうて) = kaite (かひて). Note how we now say omou and write おもう instead of omoo, and kau / かう instead of koo too. As for the -u gozaimasu pattern, it is still very occasionally used. It is a very polite way of saying desu, e.g. うれしゅうございます = うれしいです. You can sort of link this to the fact that the negative is うれしくない, the positive would, if you invert the ない, be うれしくある (which would only be used if there is a reason to), now change うれしく to うれしゅう and ある to ございます. Fifty to a hundred years ago, this formation was quite a common thing. -- KittySaturn 08:57, September 2, 2005 (UTC)

As a side note of little significance, perhaps, but maybe interesting, the おはす mentioned above, owasu (not ohasu), was the only other suru (su) irregular verb besides suru (su), which can also mean aru (ari), which is why it is also considered a possible origin of desu, which died out, which (... etc.) so now we are left with suru. :-) Not only this, but owasu is the origin of the word gozaimasu. おはす (御座す) => read kanjis in on-yomi: ござ => add verb あり (= modern day ある) : ござあり => make it yodan type : ござある => shorten it : ござる => add ます : ござります => apply onbin : ございます. There are many steps to this final gozaimasu, and between each step there are many branches (e.g. the go came off, or, gozarimasu morphed into gozansu, etc. etc.), but only one form survived all these times. -- KittySaturn 09:05, September 2, 2005 (UTC)

Oversimplified Romaji

Obviously this is not a comprehensive guide to pronounciation, but I feel the notes do not justly push the fact that Japanese simply isn't pronounced the way it looks (as Romaji). Particularly the over-simplified explanation of the pronounciation of the /o/ vowel mentioned above. Seems to be understood that the Roman spelling is merely representative and can't be relied on for accurate pronouncation (although pronouncing Japanese as it looks in English will probably render a speaker at least mostly intelligible to most Japanese) but while most basic Japanese langauge texts try to imply /wo/ changes to /o/ (as far as relative pronounciation) this simply isn't true in the most common dialects. /wo/ does undergo an audible change in MANY or MOST of Japanese speakers -- probably less predominantely in the younger generation [speculation here] -- but it is dependant on the surrounding sounds as opposed to the grammar function. For example; a fairly common name, Kaori, becomes rather uncommon when the character /wo/ is used instead of /o/. There is no change in pronounciation, even though it may be spelled Kawori. Also notable is how it sounds in music, especially when it is drawn out; the /wo/ style sound is much more pronounced. The basic reason for this is that the sound is formed in a slightly different part of the mouth (or a slightly different way) than it is in English. Thus the /wo/ sound often finds it's way into borrowed or commonly 'katakana-ized' words or phrases, for example 'carry on' which sounds like 'carry won' or 'carry uon' coming from the mouth of most non-English speaking Japanese. A very similar distinction is seen in the sounds for /n/ (for example, the pronounciation of yen vs. en), /zu/ and all the /z/ sounds (which sound more like /dzu/ etc.), and /fu/, which really isn't anything like the english /fu/ at all.

This is all very wordy and I don't believe it's exactly useful in this document, but I think it should be stressed that English romanizations for Japanese sounds are not extremely representative. At least then some people would stop argueing about which English character 'it' sounds like, knowing that it doesn't really matter at all. Gavin 2005 07 15

there's an error of translation near: "Nihon" (にほん) can mean "two books" (二本) as well as "Japan" (日本). 本 is the counter for long, thin cylidrical objects, not for books. komuta, 2005 07 21

Thank you for your suggestion! When you feel an article needs improvement, please feel free to make whatever changes you feel are needed. Wikipedia is a wiki, so anyone can edit almost any article by simply following the Edit this page link at the top. You don't even need to log in! (Although there are some reasons why you might like to…) The Wikipedia community encourages you to be bold. Don't worry too much about making honest mistakes—they're likely to be found and corrected quickly. If you're not sure how editing works, check out how to edit a page, or use the sandbox to try out your editing skills. New contributors are always welcome. — Gwalla | Talk 04:29, 24 July 2005 (UTC)

The pronunciation issue seems moot to me. Use IPA consistently and it should fix itself. Surely there must also be native speakers with decent recording equipment who can add sound files. Some minimal pair excercises would be nice, for one thing.
Peter Isotalo 16:03, 25 July 2005 (UTC)

Relationship to Chinese

Concerning the latest tweaks to the lead, could someone tell me what linguists that actually claim that Chinese and Japanese are related? If it's only completely fringe scholars, I don't see the merit of stating that it's "largely accepted". Is this on the same level as Nostratic languages, or are there reasonable doubts accepted by a majority of scholars? Personally, I've never even heard crackpot claims of Chinese and Japanese being distantly related.

Peter Isotalo 18:36, 25 July 2005 (UTC)

It's even worse than Nostratic, since there are Nostratists who include Japanese as a candidate for nostratic, but not Chinese. Tomer TALK 22:35, July 25, 2005 (UTC)

I think it's likely that the majority of linguists discount a connection between Japanese and Chinese, a judgement that seems justified based on the fact that basic aspects of their grammar systems are completely different (SOV and SVO etc.) but many (normal) people would still argue that their is a connection, due to the similarity in sound of a large number of words. It may be that Japanese changed slightly to become more similar to Chinese more recently, but I'll leave it to the linguists to even give that much credit. freshgavin 20:32, 31 July 2005

No, it's not because of the word order that all mainstream linguists reject the idea of a demonstrable genetic relationship between Japanese and Chinese. It's because there are no regular sound correspondences in the basic vocabulary. (Unfortunately, regular sound correspondence is not defined on the historical linguistics page, and I don't have the time to rectify that shortcoming now.)--Ph.D., historical linguistics
How much of these "similarities of sound" remain if you remove all (likely) loanwords?
IIRC, Chinese has CV(C) syllables where the final C is a set of different consonants dependent on dialect. Japanese in contrast has strictly CV(V)(n,Q) syllables, where Q is consonant gemination or occasionally a glottal closure. Also, native Japanese words are almost entirely two, three, or four syllables in length with few long (double) vowels, whereas Chinese loanwords are nearly all two syllables with frequent long vowels. So if you dropped the Chinese loanwords Japanese would sound very different. Just as if you dropped all the Latin and French derived vocabulary in English, modern English would sound much closer to its Germanic brethren. Removing Chinese loanwords from Japanese is left as an exercise for the reader. — Jéioosh 20:01, 5 September 2005 (UTC)

There are a few hypothesized changes from Old Japanese through Classical Japanese which happened under the influence of Chinese. For instance, the introduction of the nasal syllabic n is probably due to a reduction of final mu syllables. This may have occurred because of the common nasal n ending many Chinese words. Note that until the kana spelling reform in the 20th century there was no sign clearly differentiating mu and n. There are a few others but they aren't at the top of my thoughts right now. Other changes certainly included grammatical changes from the introduction of vast numbers of lexical items. — Jéioosh 01:02, 1 August 2005 (UTC)

Many scholars reject the idea that Chinese influenced Japanese pronunciation to any significant degree. All the changes from Old Japanese to the modern language can be explained internally, without reliance on Chinese. Also, Old Japanese was perfectly capable of adapting Chinese words without changes to its (i.e., OJ's) phoneme inventory or syllable structure.--Ph.D., historical linguistics
It's not true that there was no sign differentiating mu and n until the 20th century. It was only the case for the auxiliary verb む that was written as mu and pronounced as n, e.g. before reforms おほからむ => after reforms おおからん. And this once-extremely-common auxiliary verb is rarely used nowadays anyway. In all other places n had its own symbol for a long time. -- KittySaturn 08:39, September 2, 2005 (UTC)
I don't have my Old & Classical Japanese references with me, so I can't confirm this, but I have the suspicion that your claim is only valid for Medieval through Modern Japanese. I'd like to see a reference on this, if you have it. — Jéioosh 20:01, 5 September 2005 (UTC)

ぢゃ ぢゅ ぢょ and some others

When are the hiragana ぢゃ (ja), ぢゅ (ju), and ぢょ (jo) used in place of じゃ ja, じゅ ju, and じょ jo? Hiragana doesn't explain. Guessing from how chi and tsu to ji and zu when used after chi and tsu or a kanji relevant to the meaning of a word, is it ぢゃ if ぢゃ comes after ちゃ?

I'm trying to learn Japanese, using hiragana as my first lesson. Toothpaste 11:24, 18 September 2005 (UTC)

As for me, in fact, no word comes to mind that utilizes ぢゃ ぢゅ ぢょ, except for some slangy expressions. In most cases we use じゃ じゅ じょ.
In addtion, there're few words that inculde ちぢ (ちぢむ "to shrink" and its derivatives may be the only major example) although for つづ, there seem to be some: つづく "to continue", つづり "spelling" and つづみ "hand drum".
BTW, if you're intersted in learning Japanese, why don't you take a look at (and join perhaps?) wikibooks:Japanese? There's also a mailing list for it. - Marsian / talk 05:13, 19 September 2005 (UTC)

Thank you. By the way, is こおこお-せい the correct hiragana for "kookoo-sei"? Is だいがく-せい the correct hiragana for "daigaku-sei"? Toothpaste 06:42, 19 September 2005 (UTC)

Or is it こうこうせい for high school? The former brings up more Google results. Looks like this romaji-hiragana converter was a little off. Or was "kookoo-sei" not the right romaji? "Koukou-sei" brings up more Google results, too. Toothpaste 08:48, 19 September 2005 (UTC)

It's koukou-sei kōkō-sei (こうこうせい), which btw means "high school student". -- ran (talk) 10:24, 19 September 2005 (UTC)

Thank you very much. This link confused me with the spelling. By the way, how is "kootoogakkoo," according to that link, spelled in romaji and hiragana? Toothpaste 10:32, 19 September 2005 (UTC)

Hiragana: こうとうがっこう
Hepburn Romaji: Kōtōgakkō
What I type into Windows: koutougakkou
The website that you showed me: kootoogakkoo
The thing is, there are several types of Romaji, and Hepburn (the most common one) doesn't distinguish おお from おう. That's one reason why it's advisable to start thinking in terms of kana and kanji instead of romaji as early as possible.
-- ran (talk) 10:39, 19 September 2005 (UTC)
Here comes edit conflict... But I do not mind (well, I'm a lazy man ;)
  • The hiragana for 高校生 (high school student) is こうこうせい. But it is pronounced as こおこおせい or, more likely, こーこーせぇ. だいがく せい is for 大学生 (university student). Here's my voice for you: こうこうせい and だいがくせい.
  • According to Rōmaji, there're four or more romanizetion methods for Japanese. Since I use romaji just for typing and some, I don't know much about the systems, but perhaps either "kōkōsei" (Hepburn, standard for en.wikipedia) or "koukousei" (Wāpuro) is relatively common. According to Hepburn romanization#Long vowels, "kookoo-sei" seems to be "modified Hepburn".
  • I tried your converter a little. Probably it uses Wāpuro romaji. Actually, when I write 高校生, I type "koukousei" then push space-key to convert it into kanji via IME. For 東京 (とうきょう), you have to type "toukyou", not "tokyo" (please try it for the converter. "tokyo" becomes ときょ, and it's correct for "Wāpuro" romanization).
Please see the links and Rōmaji for more. - Marsian / talk 11:05, 19 September 2005 (UTC)

Grammar section

The grammar section of this article is becoming extremely long - much too long for a summary of the main article. I'd like to suggest shortening it very drastically to give the basics, and moving any non-duplicated contents from here into "Japanese grammar". If there are any objections, please state your case. --DannyWilde 02:24, 20 September 2005 (UTC)

I think the grammar section is still somewhat too long. There's too many examples, bullet lists and repetitions of the same aspects. Try to avoid Japanese jargon. The grammar maybe different from a lot of Indo-European languages, but that doesn't mean that an i-adjective should be called "keiyoushi". Translate them once and then stick to English. The article is intended for more than just the grammar otaku.
The honorifics could definetly use a trimming considering they have two separate articles already. Quite comprehensive ones at that.
Peter Isotalo 15:52, 15 November 2005 (UTC)

Major removals to be made

Further to the above comments, prior to actually doing the edits, here I have copied out the grammar section of the article, and written my comments in italics and suggestions for material to be moved elsewhere, simplified, or deleted, which are written like this. Please keep in mind that this section should be a summary of the main article Japanese grammar when commenting - I am not suggesting removing this material from Wikipedia altogether, but from the grammar summary in the top page on the Japanese language. If anyone has any points to make, please let me know and I'll try to keep it in mind. --DannyWilde 05:14, 5 October 2005 (UTC)

Certain aspects of Japanese grammar are highly controversial. Japanese grammar can be characterized by the following prominent features:

  • The basic sentence structure of a Japanese sentence is topic-comment. For example, consider the sentence kochira wa Tanaka san desu こちらは 田中さんです。.

Remove all kanji/kana. They do absolutely nothing to illustrate Japanese grammar, and serve to make the section unreadable. Romaji is enough for the grammar section. --DannyWilde 05:14, 5 October 2005 (UTC)

Kochira こちら is the topic of the sentence, indicated by the particle wa は (Note: ha は is pronounced wa when it is a "topic marker"); literally, it means "as for this direction," but here, it means "as for this person." The verb is desu です ("be"). As a phrase, Tanaka san desu 田中さんです is the comment. This sentence loosely translates to "As for this person, (it) is Mr./Mrs./Ms. Tanaka".  So Japanese, like Chinese and Korean, is often called a topic-prominent language, which means it indicates the topic separately from the subject, and the two do not always coincide. For example, the sentence Zō wa hana ga nagai 象は 鼻が 長い literally means, "as for elephants, the nose is long." The topic is zō 象 "elephant," and the subject is hana 鼻 "nose."
  • Japanese nouns have neither number nor gender. Thus hon 本 "book" can be used for the singular or the plural. However, it is possible to explicitly indicate more than one, either by using numbers or by using certain forms that refer to groups. There are several noun suffixes that indicate groups; the most common are -tachi, -ra, and -domo. These collective suffixes are not true plurals; rather, they indicate groups, one member of which is chosen as representative. Their use is optional, and restricted to animate beings. Furthermore, their exact meaning is contextual: oneesan-tachi お姉さん達, literally "big sister (and her) group," could refer to someone's older sisters; an older sister and her friends; or an older sister and her family. Another example is Saitō-san-tachi. This does not refer to a group of people named "Saitō"; rather, it refers to a group of people that includes at least one person named "Saitō," effectively meaning "Saitō and her party," or "Saitō and the rest of them." Also, there are a small number of native words that indicate the collective through reduplication. For example, hito 人 means "person" while hitobito 人人/人々 means "people"; ware 我 is an archaic form of "I," while wareware 我我/我々 means "we", specifically the exclusive "we".

Rationale: This duplicates Japanese grammar and it is not important enough to repeat in the summary.--DannyWilde 05:14, 5 October 2005 (UTC)

  • Verbs are conjugated to show tenses, of which there are two: past and present (also called non-past tense, since the same form is used for the present and the future). The present tense in Japanese serves the function of the simple present and the future tense, while the past tense (or perfect tense) in Japanese serves the function of the simple past tense. The distinction is between actions which are completed (perfect) or are not yet completed (imperfect). The present perfect, present continuous, present perfect continuous, future perfect, future continuous, and future perfect continuous are usually expressed as a gerund (-te form) plus the auxiliary form imasu/iru. Similarly, the past perfect, past continuous, and past perfect continuous are usually expressed with the gerund plus the past tense of imasu/iru. For some verbs, that represent an ongoing process, the -te iru form regularly indicates a continuous (or progressive) tense. For others, that represent a change of state, the -te iru form regularly indicates a perfect tense. For example, kite imasu 来ています regularly means "I have come", and not "I am coming", but tabete imasu 食べています regularly means "I am eating", and not "I have eaten". Note that in this form the initial i of imasu/iru is often not voiced, especially in casual speech and the speech of young people. The exact meaning is determined from the context, as Japanese tenses do not always map one-to-one to English tenses. In addition, Japanese verbs are also conjugated to show various moods.

Rationale: This material is not important enough for the summary. It can be summarized in a few sentences and non-duplicated details moved to Japanese grammar. Further, it is not very accurate. --DannyWilde 05:14, 5 October 2005 (UTC)

  • There are three types of words that can be said to correspond to adjectives in English: 形容詞 keiyōshi, referred to variously as stative verbs, adjectival verbs i-adjectives, or simply adjectives (the Japanese word keiyōshi is used to denote an adjective in English); 形容動詞 keiyōdōshi, called copular nouns, na-adjectives, adjectival nouns, or quasi-adjectives; and , 連体詞 rentaishi, called true adjectives, prenominals, or pre-noun adjectivals. Both keiyōshi and keiyōdōshi may predicate sentences, and both can be said to inflect, though they do not show the full range of conjugation found in true verbs. There are regular ways to turn the both keiyōdōshi and keiyōshi into adverbs. The rentaishi are few in number, and unlike the other words, are limited to modifying nouns. Adjectives never predicate sentences. Example include ookina 大きな "big" and onaji 同じ "same" (although there is a noun 同じ onaji that can be used as a predicate, as in 同じだ onaji da). It is worth noting that because the widespread study of Japanese is still relatively new in the Western world, there are no generally accepted English translations for the above parts of speech, with varying texts adopting different sets, and others extant not listed above.

Rationale: This material is very detailed and useful, and certainly should be preserved somewhere. However, it is much too detailed for the summary. I suggest moving into a new page on Japanese adjectives or into Japanese grammar. --DannyWilde 05:14, 5 October 2005 (UTC)

  • The grammatical function of nouns is indicated by postpositions, also called particles. These include possession (no の), subject (ga が), direct object (o を) (Note: wo を is pronounced o when it is used as a direct object marker), indirect object (ni に) and others. The topic is marked by the particle (wa は). These particles play an extremely important function in Japanese, though some can be elided in casual speech.

Minor detail. --DannyWilde 05:14, 5 October 2005 (UTC)

*Japanese has many ways to express levels of politeness. These strategies include the use of special verbal inflection, the use of separate nouns and verbs indicating respect or humility, and certain affixes.

Repetition of subsection. --DannyWilde 05:14, 5 October 2005 (UTC)

*The word desu/da is the copula verb. It does not correspond exactly to the English be verbs, and often takes on other roles. In the sentences above, it has played the copulative function of equality, that is: A = B. However a separate function of "to be" is to indicate existence, for which the verbs aru ある and iru いる are used for inanimate and animate things, respectively.

Rationale: simplify to two or three sentences --DannyWilde 05:14, 5 October 2005 (UTC)

  • The verb "to do" (suru, polite form shimasu) is often used to make verbs from nouns (aisuru "to love", benkyō suru "to study", etc.). Japanese also has a huge number of compound verbs (e.g. tobidasu "to fly out, to flee," from tobu "to fly, to jump" + dasu "to go out").

*Japanese has many words that are translated as pronouns in English. However, none of these words are grammatically pronouns in Japanese, but may be thought of instead as referential nouns. Referential nouns are all regular nouns, in that they may be modified by adjectives, whereas true pronouns may not be. For example, a Japanese speaker can say manuke na kare wa nani mo shinai "stupid (copula) he (topic) nothing does", but in English this would have to be broken into two statements, as we cannot say "stupid he": "he's stupid and doesn't do anything". Which one of these referential nouns is used depends upon many factors, including who is speaking, who is being spoken to, and the social setting. Their use is often optional, since Japanese is described as a so-called pro-drop language, i.e., one in which the subject of a sentence does not always need to be stated. For example, instead of saying Watashi wa byōki desu "I am sick," if the speaker is understood to be the subject, one could simply say Byōki desu "To be sick." A single verb can be a complete sentence: yatta! "(I / we / they / etc) did (it)!".

Rationale: Repetition of material in Japanese grammar. Too detailed for summary. I suggest simplifying this and moving any non-duplicated material elsewhere. --DannyWilde 05:14, 5 October 2005 (UTC)

Languages of the Three Kingdoms, and relations

How were the languages of the Three Kingdoms of Korea related to each other, and what is their supposed relationship to modern Korean and Japanese. Wikipedia doesn't seem to be particularly clear on this subject.

long see also list

the see also list was flagged as long. as many of the entries in the list are in the category Japanese Language, shld they be removed from the see also list, while only retaining a link to the category?

following are the links in the JapLang see-also list also present in JapLang category.

  1. Four-character idiom
  2. Henohenomoheji
  3. Japanese language and computers
  4. Japanese literature
  5. Japanese pitch accent
  6. Japanese pronouns
  7. Japanese proverbs
  8. Japanese Language Proficiency Test
  9. Kanji kentei
  10. List of Japanese learning resources
  11. Rendaku
  12. Japanese titles

Doldrums 19:18, 12 October 2005 (UTC)

Is it possible to check "What links here" for those pages? For example does "Henohenomoheji" have any links except the category and the Japanese language page? "Rendaku", "Kanji kentei", are well linked pages, and "Japanese titles" is linked from other another place in Japanese language. I'll chop out some obvious bad guys and then if it is still a problem, please discuss here again. --DannyWilde 22:28, 12 October 2005 (UTC)

Remove kanji/kana from grammar section

I already made the point above: Remove all kanji/kana. They do absolutely nothing to illustrate Japanese grammar, and serve to make the section unreadable. Romaji is enough for the grammar section. At that time, I removed all the kanji and kana from the grammar section. I still don't see any sensible reason to add the kanji and kana to the grammar section, hence I removed it all again today. If there is a good reason to have it, please discuss here. --DannyWilde 05:22, 15 November 2005 (UTC)

FWIW, I support. Within language-related articles, kanji and kana should be used only in discussions about the writing system itself. --Pablo D. Flores (Talk) 10:26, 15 November 2005 (UTC)
Oppose. I find that with most languages -- and Japanese is no exception -- the original script is easier to read than transliteration for anything more than a word or two. Far from making the section "unreadable", I think that the presence of Japanese writing actually improves readability. I think the transliteration should stay, but having only transliteration looks silly to my eye. --Marnen Laibow-Koser (talk) (desk) 12:42, 15 November 2005 (UTC)
You argue for keeping it on the basis that you find the kanji and kana easy to read. OK, but do you need to read an elementary summary of Japanese grammar? No, we can be sure that you don't, because you can already read kanji and kana. However, this grammar summary is meant for people who know very little about Japanese. The kanji and kana do not belong in that grammar summary. --DannyWilde 21:59, 15 November 2005 (UTC)
I agree with Danny. Marnen, while I agree that there are many places where including the native script is beneficial, I just don't see how kanji and kana is "potentially useful information" for the type of person who'd be reading this grammar section. Sometimes too much information really is too much information. CES 22:39, 15 November 2005 (UTC)
Danny, you're arguing from false premises. Your argument could be restated this way:
(1) Premise: Marnen has a good knowledge of kanji and kana.
(2) Premise: From (1), we can tell that Marnen has advanced knowledge of Japanese.
(3) Conclusion: Given (1) and (2), Marnen is not the target audience for this article.
However, as it happens, neither (1) nor (2) is true, and since your premises are false, your argument is logically invalid. I have a small knowledge of kanji and kana (which is why I only put them in a few places and requested help from those who know more than I do). Likewise, I have a mediocre knowledge of the Japanese language. I read this article to get more information on the Japanese language, so I am very much a member of this article's target audience. And as a member of this article's target audience, I strongly suggest that the Japanese script is an asset.
I think it pays to look at other similar Wikipedia articles, too. In Russian grammar, all the Russian is in Cyrillic script. Likewise for Greek language: all the Greek is in Greek script (with Roman or IPA equivalent as necessary). This is the usual method of dealing with such articles, both within Wikipedia and elsewhere. I see no reason why the situation should be any different for Japanese than it is for Greek, unless there's a Eurocentric bias happening here.
Finally, I'd ask this: does the presence of the Japanese script harm the article? (I'd answer that question in the negative.) If it is not hurting the article, then it should stay. --Marnen Laibow-Koser (talk) (desk) 19:49, 16 November 2005 (UTC)
Agree. I think that for people not that familiar with Japanese would have an easier time reading the transliteration as they can actually read it, the average person cannot read the Kana/Kanji and hence the sentences blur into nothingness for them. Ben W Bell 12:44, 15 November 2005 (UTC)
Agree - in this case the kanji/kana is at best eye candy and at the worst a distraction for the majority of people, who can't read it. For those who can read it, romaji is more than enough to illustrate these simple grammar points. CES 13:41, 15 November 2005 (UTC)
Agree on that too. I can read kana and simple kanji, but romaji is perfectly OK for examples, especially since this is English Wikipedia; if I wanted to practice reading I'd try the Japanese version. The Manual of Style recommends using romaji for all words in Japanese including names (clarifying with kanji/kana at the beginning if needed, at most), and warns that many users do not have CJK fonts installed. I do, but still most kanji look hideous. --Pablo D. Flores (Talk) 13:56, 15 November 2005 (UTC)
Marnen, you're not taking into account that people who don't know a single character of Japanese are supposed to be able to read this. I've reverted all kana and some kanji and I agree with everyone else that romaji should be used unless there's a very good reason to use kana or kanji.
Peter Isotalo 15:35, 15 November 2005 (UTC)
Please don't assume that you can read my mind. In fact, I am taking into account that people who can't read Japanese are supposed to be able to read this -- that's precisely why I think keeping the transliteration is essential. However, I still argue for keeping the Japanese writing. People who can read it will find it helpful, and people who can't read it can just skip over it. The reverse approach essentially dumbs it all down to the lowest common denominator, and thus penalizes people who want the additional info.
I should also mention that I think Pablo's citation of the Manual of Style is irrelevant here. While most articles on things Japanese should indeed not include much Japanese script if any, I think this article is in a somewhat different position, since it is supposed to be a detailed treatment of the Japanese language. As such, it should include as much Japanese-language-related information as possible -- including words in their native script when needed.
To those who still want to delete the Japanese script, I say this: you are deleting potentially useful information from an encyclopedia. I believe that we should generally err on the side of giving too much information (so the reader can sort it out) rather than giving too little (and watering down our raison d'être). --Marnen Laibow-Koser (talk) (desk) 20:34, 15 November 2005 (UTC)
Yeah, great, let's turn all the Wikipedia articles into unreadable gibberish by piling in every little piece of unimportant linguistic trivia we can into the summaries. Our motto must be that every little piece of trivia and kanjicruft is too precious to remove, especially from a summary. If the summary is short, to the point, relevant and readable, then it might be useful to a lot of people, and we can't have that: article summaries are meant to be overlong lists of disorganised trivia and cruft, aren't they?
There are extensive sub articles like Japanese grammar and Japanese writing system where all the details will fit in nicely, but why bother editing carefully when the summary is there right at the very top to absorb every detail? After all, putting the information in the correct place involves the tedious task of actually reading the existing articles, thinking about the overall structure and readability, and we wouldn't want these articles to become readable, would we? Much easier and better to turn them into utter gibberish.
Over the time I've been editing Wikipedia, I saw the article katakana go from being fairly good, well written, well informed article to being wikifiddled into a piece of unreadable crap because editors were adding their little bits of trivia to the summary at the top of the article. The trivia wasn't wrong, and it might even have belonged in the article, in the right place, but for some reason the editors insisted that every single tiny detail had to go right at the very top. Incidentally I've worked over it since then, but quite recently someone added to the list of "main uses of katakana in modern Japanese" at the top of the article, that it was used for shakuhachi music notation. Right, that's obviously a "main use". Incidentally, there was a big list of uses just under the contents where that music information could have gone, but no, every Wikifiddler lives by the insane motto "I must add information to the summary! I must make this article more and more unreadable!", thus the use for writing shakuhachi music is added as "one of the main uses of katakana" rather than another minor use. Is it possible to plead for just a tiny modicum of common sense in these discussions? Kanji and kana clearly do not belong in this grammar summary. --DannyWilde 22:37, 15 November 2005 (UTC)
No one was talking about editing the summary. --Marnen Laibow-Koser (talk) (desk) 19:16, 16 November 2005 (UTC)
On rereading it, I see that you were using "summary" to refer to this summary article, not (as I had thought) the summary paragraph at the top of the article. And so your comment deserves more of a response than I had first thought. I would say this: the problem of wikifiddling is not that there is too much information, but rather that the information is too poorly organized to be usable. That's certainly a problem in many articles, but I don't think it's at all related to what we're discussing here. --Marnen Laibow-Koser (talk) (desk) 19:52, 16 November 2005 (UTC)
Wikipedia is not helped by adding character codes that show up as white boxes or question marks for a lot of users, Marnen. Those who can read kana know perfectly well how to convert it to romaji and vice versa. Keeping the kana is just there to spoil the (most likely non-Japanese) Japanese-otakus. It's not useful information, not even potentially.
Even if I think Danny is ranting a tad, I agree that one of Wikipedia's biggest problems right now is some article gathering too much useless material. We're supposed to be encyclopedists, not information pack rats and some additions just clutter up articles needlessly. Hoarding information is not the way to write good articles.
Peter Isotalo 01:05, 18 November 2005 (UTC)
Peter, I appreciate hearing your viewpoint, but I really don't think you're taking all the facts into account. Let me respond individually to some of your assertions that I disagree with.
"...character codes that show up as white boxes...for a lot of users..."
For better or for worse, Wikipedia uses Unicode, and editors are generally encouraged to take full advantage of the character set. These days, practically every computer I see out there can display Unicoded Japanese. Certainly there are still some that can't, but that's an argument for keeping the romaji, not deleting the Japanese.
"Keeping the kana is just there to spoil the (most likely non-Japanese) Japanese-otakus. It's not useful information, not even potentially."
Name-calling and labelling is not necessary here, and only serves to cloud the issue -- for example, I'm not sure what you mean by "Japanese-otaku" (although I will tell you that I'm not your stereotypical gaijin Japanophile). I have studiously tried to avoid such things throughout this discussion, and I would ask you to do the same. At any rate, the kana (and kanji) certainly are useful information, at least to some subset of readers (such as myself). Just because you don't find them useful doesn't mean they can't be useful to anyone. Again, if this were a question of Cyrillic script on a page about Russian, I don't think that we'd even be having this debate. Why should Japanese be any different?
"Hoarding information is not the way to write good articles."
If by "hoarding", you mean "presenting disorganized infodumps", then I agree with you 100%. But I think that presenting a language in its native script is about as far from a disorganized infodump as it is possible to get. If you don't like the Japanese script, you are free to ignore it, but don't keep it away from those of us who find it useful. That's kind of tyrannical, IMHO, and contributes to what may become a gradual dumbing-down of Wikipedia (I've seen it threaten to happen in other articles too). Remember, we are not a paper encyclopedia, and so we should not feel confined solely to the scope that paper-encyclopedia articles would have. --Marnen Laibow-Koser (talk) (desk) 20:02, 18 November 2005 (UTC)
What would be convincing to me is to hear exactly how having the script in this particular case is useful. I'd also like to hear how not having the script dumbs down the article. What does the Japanese script add that the romaji doesn't, besides the answer to the question "How do you write (insert phrase here) in Japanese?" Information for information's sake is called trivia. Too much trivia and the article becomes trivial, pun fully intended. CES 21:23, 18 November 2005 (UTC)
Good questions, CES. In fact, these are the very questions I was trying to answer above at length. However, if my longer answers above were not sufficiently clear, here's a brief version.
How is it helpful to have the Japanese script in the article? 
For those of us who read even a little Japanese script, it is generally less awkward to read the language in the Japanese script than in transliteration. This is exactly the approach taken in articles on Russian and Greek, and I think it's the appropriate approach for any language. Of course, it's also essential to keep the transliteration, so that people who don't read the native script can still understand the article.
This is just a style issue. If the Japanese script added information rather than making it "generally less awkward" for you, I might be persuaded. CES 02:06, 22 November 2005 (UTC)
How does removing the Japanese script dumb down the article? 
By gearing the article to the lowest common denominator (i.e., the people who don't read Japanese script). I have no problem with catering to the lowest common denominator; my problem lies with not providing an alternative presentation for people who want such a thing. That forces everyone to the lowest-common-denominator version, and so lessens the information content of Wikipedia.
Why are people who don't read Japanese "the lowest common denominator" when the question is not knowledge of the writing system, it's knowledge of Japanese grammar? If you want a comprehensive explanation in Japanese, why not go to the corresponding Japanese Wikipedia page? Or even, the more comprehensive Japanese grammar article? I don't see why we should cater to people who read Japanese at a beginning-to-intermediate level in a summary outline on grammar. CES 02:06, 22 November 2005 (UTC)
Information for information's sake is called trivia. 
If that is how you define trivia, then you would seem to be saying that the entire Encyclopædia Britannica is nothing but thousands of pages of trivia.
If the entire Encylclopedia Britannica was translated sentence-for-sentence into Japanese for the sake of native English speakers, it would indeed be trivia. CES 02:06, 22 November 2005 (UTC)
Finally, I have a couple of questions for you (or anyone else in this debate who feels like stepping up to the plate).
  1. Would you object in the same way to having Cyrillic script in an article on the Russian language?
Not if it adds information to the article. I have yet to hear what information is being added in this case. CES 02:06, 22 November 2005 (UTC)
  1. How is the presence of the Japanese script harming the article? In other words, why fix it if it ain't broke? --Marnen Laibow-Koser (talk) (desk) 21:08, 21 November 2005 (UTC)
Honestly, if Danny hadn't removed it, I probably wouldn't have said one thing or another about it being there. But now that it's become a keep/delete issue, the argument for removing it was stronger than the argument for keeping it. This reminds me of a while back when we had someone who was passionate about the Kunrei system of romanization. He wanted to have Hepburn, Kunrei, and I think either JSL and/or Nihon-shiki present whenever a term or sentence was romanized. Would having this information "harm" the articles? Some would argue that having the alternate romanizations might help people. But the overall effect is that it clutters the article and makes it difficult to read. It's a cumulative effect. Today we add the Japanese script, tomorrow we add the sentence in just hiragana because some people can't read kanji, and the next day we add another romanization system. Where do we draw the line? I say we keep things as simple as possible. If you can't make a positive argument for keeping article content, maybe this sends a strong signal that the content is not really needed.
Another problem that I think we have is what Peter refers to with his "otaku" statement. Many times I feel that articles are written by and/or cater to beginning/intermediate students of Japanese who feel the need to demonstrate their Japanese ability. People who know enough to be able to read kana and a few kanji but not enough to be able to read primary source material. This spawns a lot of irrelevent "Japanese lessons" within articles, often with incorrect information. I hope Danny doesn't mind me linking this, but I think he makes good points on his user page so I won't repeat them: [[3]].
Don't get me wrong, I think there are many times when having the language's native script is useful. But I think its usage needs to be justified in an English language encyclopedia. "What does it hurt?" doesn't cut it. If it can't be justified, why add it? CES 02:06, 22 November 2005 (UTC)
You're certainly making me think harder about the points I'm trying to make. I wonder, though, why you're saying that including the J is "just a style issue" and therefore not worth doing: seems to me that if including the J is a style issue, deleting the J is equally one. You claim that making the article less awkward to read is not an appropriate reason for including the J, yet you then cite awkwardness as a reason for deleting the J. I don't think it's fair to have it both ways.
I think this is an example of a larger problem. I don't want to whine, but it's increasingly looking to me like every time I give a reason for including the J, it is brushed off with scarcely a nod, while the reasons given by you and others for deleting the J are (IMHO) flimsy and, in some cases, even use the same arguments that are being derided as insufficient reason for including the J -- many of them seem to boil down to "well, I can't read the J, so it must not be necessary" or some such. (I really hope I'm wrong, but that's the way it looks to me right now.) Either that, or we're talking past each other at cross-purposes. There's certainly a disconnect here, and I'm not sure how to bridge it.
Again, I say to all who are against keeping the J: just because you can't read it doesn't mean it shouldn't be in the article, and by the same token I'm not just arguing for keeping it because I can read it. As a matter of principle, I think languages belong in their native script wherever possible. (If I were editing an article on, say, the Limbu language -- whose script I read not at all -- I'd still be arguing for the inclusion of the Limbu script. Seeing how the language is actually written is undeniably a Good Thing, IMHO.) --Marnen Laibow-Koser (talk) (desk) 15:32, 22 November 2005 (UTC)

For me it's a question of style and content. In this case, I think the script adds no content and has a marginally negative impact on style. If you could convince me that there's even the slightest content-oriented benefit to keeping the script, I'd probably argue for its re-inclusion. But my inference from the replies on this subject is that most people either can't read the script so it adds nothing or they can read the script but it still adds nothing that the romanization doesn't. Count me in the latter category: I see no content or style benefit to adding the script.

Personally I think it all should be a moot point. The grammar section is way too long as it is, given the nice comprehensive Japanese grammar article. I'd support Danny's proposal (above) to trim down the grammar section to bring it more in line with the other languages' articles (e.g. French language, Spanish language). CES 16:55, 22 November 2005 (UTC)

Grammar Glosses

Does anyone besides me find the glosses to the following phrases in the grammar section confusing?

"Negatives are formed from verb conjugations. For example,

   大学にいく。 Daigaku ni iku. "Go to university.",

becomes

   大学にいかない。 Daigaku ni ikanai. "I don't (he doesn't) go to university."　(Indicative, not imperative.),

with iku "to go" changing to the negative form ikanai."

If I didn't understand the Japanese phrases, I'd really have little idea of what exactly is going on in these sentences. It seems like part of the problem is in the translation itself (the first phrase is not even a complete sentence and sounds like an imperative, the two translated phrases aren't really parallel in meaning, and the "(he doesn't)" in the second phrase is certain to confuse many people). Part of the problem is that the word "university" is used in a non-standard manner for US English at least. Finally, part of the problem is the fact that the phrases can indicate going to school in the general sense (as in "she goes to school at State U.") or in a specific sense ("she went to school today").

It seems like the point of this section should be to simply show how "X (Verb)s" becomes "X doesn't (Verb)". How about replacing the above example with something simpler like:

Pan o taberu. "I will eat bread."

Pan o tabenai. "I will not eat bread."

I realize these sentences could be translated different ways, but for our purposes as an intro to Japanese grammar, doesn't something like this make more sense? If someone has a better example, that would be great. But the phrases we have now are downright confusing. CES 14:25, 15 November 2005 (UTC)

It's been Wikifiddled since I wrote the original example, but admittedly it wasn't very good originally. I'd suggest sticking to simple things here; the point is to illustrate how the negative is formed rather than to get into future tenses. --DannyWilde 22:43, 15 November 2005 (UTC)
I went ahead and changed the example. CES 01:41, 17 November 2005 (UTC)