Talk:Japanese grammar/Archive 4

Archive 1

Archive 2

Archive 3

Archive 4

Word class system

Is there any reason why adverbs and particles are not mentioned in the "Word class system" section?

Also, is there any reason why the word "lexical" is in bold in the sentence "Japanese has five major lexical word classes:"? Is it emphasising that these are lexical word classes rather than some other type of word classes? If so, italics would be better than bold, and it would be helpful to explain, if it can be done briefly, what the difference is. 86.135.171.19 (talk) 14:49, 8 January 2011 (UTC)

Head words

In the section "Word order: head final and left branching":

• noun governed by an adposition ("on the table", "underneath the table");

Does this mean that "on" and "underneath" are the "head" words? This seems peculiar to me. Naturally I would imagine "table" to be the "head" word. Is there any way a few words of explanation and justification could be added to the article? 86.174.160.123 (talk) 12:35, 10 January 2011 (UTC).

Better example available?

The example "Bob bought some flowers and Mary a vase" seems awkward to me, like a garden-path sentence. I feel a better example should be possible. How about "Bob bought his mother some flowers and his father a tie". Or is the intention that the subject should be different in each clause? 86.174.160.123 (talk) 12:45, 10 January 2011 (UTC)

I agree, this example is not optimal. I'll change it momentarily to use the example above. -- Eiríkr Útlendi | Tala við mig 20:19, 18 January 2011 (UTC)

Pronouns

In the section "Word class system", the article says:

"It is also opposite to the situation with pronouns, which are closed-class in Western languages but open-class in Japanese and some other East Asian languages."

This is implying that pronouns are a Japanese word class, but they were not included in the earlier list of Japanese word classes. —Preceding unsigned comment added by 86.174.160.123 (talk) 12:54, 10 January 2011 (UTC)

Present and future

"The semantic difference between present and future is not indicated by means of conjugation. Usually there is no ambiguity because few verbs can operate in both uses."

Is this saying that few verbs can be used to talk about both present events and future events? Is that really true? —Preceding unsigned comment added by 86.161.81.131 (talk) 21:58, 10 January 2011 (UTC)

Any verb (AFAIK) can be used to talk about both present and future events; I'm really not sure where the quoted sentence above came from, or quite what it was supposed to convey. I've since changed the wording and removed the phrase "few verbs can operate in both uses", as that seems to be just plain incorrect. -- Eiríkr Útlendi | Tala við mig 23:21, 18 January 2011 (UTC)

Imperfective form

When introducing the imperfective form, the article says it "is used for plain negative (of verbs), causative and passive constructions."

The table following this lists two forms of this "imperfective" for some verbs; for example, 書か (kak.a) and 書こ (kak.o), which are said to be in order of "increasing rarity". It's natural at this point to assume that the two different forms are used for the same purpose, but one is just more commonly used than the other. However, in the next table, it's shown that one of these forms, 書こ-, is used for the volitional "tense" (not mentioned earlier in the list of uses), and the other form, 書か-, is used in various other situations.

If it really is true that the stem of the volitional "tense" is also a type of "imperfective" (which I haven't been able to verify from any other source, though I've no particular reason to believe the article is incorrect), then I suggest that the volitional use is added to the list of uses mentioned at the top of this post, and also that "increasing rarity" is clarified or expunged as it relates to these variants (it probably is the case that 書か- forms are more common than 書こ- forms, but if they are used for completely different purposes then the existing wording is IMO misleading). 86.161.86.235 (talk) 22:41, 17 January 2011 (UTC)

This table (and indeed the whole article) could use some reorganization. "Imperfective" doesn't seem to me the right word; despite showing up in many places, most sites that I can find that use "imperfective" for 「未然形」 describe it along the lines of:

Literally, "uncompleted state form", also called Imperfective (not to be confused with the Imperfect in languages such as French), opposite of 已然形. This base is used for things not yet completed or done. (Wagawiki: The Japanese Page .com, Conjugation Bases (web), retrieved 2011-01-18)

However, this implies that the action has been started. Rather than "uncompleted" as above, 未然 means something more like "before it happens". Japanese-language materials describe the 未然形 as:

その事態が未だ起きないことを示す形という意での命名。 (大辞林 (book), 1995, ISBN 4-385-13900-8)

To wit, the action of the verb is not just imperfect (i.e. incomplete), it hasn't even begun yet. Masayoshi Shibatani (Masayoshi Shibatani (1990), The languages of Japan) and others of a more linguistic bent use the term irrealis as the English gloss for 「未然形」, perhaps to avoid this issue and any possible confusion with the imperfect and imperfective moods and modes used in other languages, where the action of the verb generally has begun but hasn't finished yet.

That aside, the volitional in modern Japanese grew out of the 未然形 form of a verb (ending in -a) plus the 助動詞 ("helping verb") -う, essentially meaning the same as the -よう / だろう / しよう used in modern Japanese to express volition or expectation. The pronunciation -au changed over time into modern -ou, and spelling finally caught up some time in the early 20th century. Novels by Natsume Soseki may still use the -au spelling, among other archaisms.

I'll see what I can do to clean up the Imperfective terminology / explanation later on. -- Eiríkr Útlendi | Tala við mig 21:14, 18 January 2011 (UTC)

FWIW, the terminal (終止形) and attributive (連体形) forms, identical in modern Japanese, roughly equate to the simple present tense in English and the imperfective aspect of languages such as Navajo, where an action is underway at some point in time (which could be past, present, or future) but has not completed. Note that this does not equate to the imperfect tense of languages such as Spanish, which is a combination of the imperfective aspect and the past tense and is thus always in the past. Note also that the action is underway, unlike the 未然形 where the action has not begun, making the use of the term imperfective for the 未然形 form quite unfortunate. -- Eiríkr Útlendi | Tala við mig 23:35, 18 January 2011 (UTC)

Yes, we need to clean this up. 未然形 is in no way imperfective; for one thing, the IPFV is a grammatical aspect, whereas the 未然形 is an inflectional stem, not an aspect at all. Therefore it is not irrealis either, though maybe half of its forms are irrealis. ("Irrealis stem" may not be a bad attempt at a translation, though the Japanese definition seems wrong, as I can't see how a completed causative can be argued to have not yet happened--maybe I'm missing something?) As you say, it's the dictionary form of the verb which is the IPFV in Japanese. — kwami (talk) 07:50, 19 January 2011 (UTC)

Replaced IPFV w IRR apart from those cases where it actually was aspect. That's an improvement; if anyone can come up with a better term, we can always change to that. Meanwhile I've purposefully not linked to irrealis, as I didn't want to encourage people to think that's what it actually is. — kwami (talk) 07:59, 19 January 2011 (UTC)

Irrealis

Thank you, Kwami. I'm curious, though -- reading quickly again through the first half of the Irrealis page, I find that the 未然形 seems to align quite well. What about the 未然形 strikes you as not irrealis-ish? -- Intrigued, Eiríkr Útlendi | Tala við mig 08:08, 19 January 2011 (UTC)

The causative and passive, each of which has their own 未然形 form. They are valence changers, but AFAIK are simply verbs, have the normal realis uses of verbs, and don't have any particular aspectual or modal meaning. — kwami (talk) 09:10, 19 January 2011 (UTC)

Ah, yes. However, the せる・させる and れる・られる of the causative and passive 1) are 助動詞 unto themselves, as you note, conjugating as 下一段 verbs, where the irrealis is essentially the same as the verb stem -- so …られ and …食べ could be expressing either the 未然形 or 連用形, and 2) could be in an irrealis mode semantically in terms of "not having happened yet" / hypothetical / subjunctive / etc. just as other verbs can. No? -- Eiríkr Útlendi | Tala við mig 19:00, 19 January 2011 (UTC)

Yes, they could be irrealis, but not because of the 未然形 inflection of the root. So that 未然形 does not make them irrealis. Thus 未然形 is very commonly not used with irrealis forms. So, it's IRR in the NEG and VOL, but not IRR (in the root) in the PASS and CAUS. "IRR" therefore only captures two of the four functions. — kwami (talk) 23:04, 19 January 2011 (UTC)

Okay, I think I might see where you're coming from -- you're looking primarily at the root, yes? I was looking primarily at the part of the verb that was conjugating, the bit at the end, be it the 動詞 itself or the 助動詞. To clarify, for the CAUS and PASS 助動詞, are you saying that there is no irrealis sense? Or, are you saying that the 未然形 of the CAUS and PASS 助動詞 does not put the 動詞 stem in an irrealis mode, semantically speaking (although it must be conjugationally speaking in order to take a CAUS or PASS 助動詞)?

If the former, I must disagree, as noted above at 19:00: a 助動詞 in the 未然形 is, pretty much by definition, 未然 -- irrealis, "hasn't happened yet". If the latter, my understanding is that the 未然形 for either the 動詞 or CAUS and PASS 助動詞 still applies to the verb as a whole -- reading over the list of implied meanings at Irrealis, all of these that apply to 動詞 seem to do so whether or not the stem verb is suffixed with a CAUS or PASS 助動詞. So 未然形の「食べ-」 or 「書か-」 are, as best I can parse it, in an irrealis mood, and likewise 未然形の「食べさせ-」 or 「書かれ-」 are also in an irrealis mood (or adverbial, if serving as the 連用形 instead of the 未然形).

Or, perhaps I've completely missed you, and instead you simply mean that a verb stem that is in the 未然形 in order to take a CAUS or PASS 助動詞, regardless of how that 助動詞 is conjugated, does not itself constitute an irrealis sense?

If so, I still think irrealis applies, more in the potential sense listed on the Irrealis page -- the core verb idea is still 未然 until the 助動詞 is applied. If I'm reading the entry correctly, -れる・られる arises ultimately from -ゆ, itself meaning 「自発。ある動作が自然に行われること」（Kokugo Dai Jiten Dictionary. Shinsou-ban (Revised edition) ｩ Shogakukan 1988/国語大辞典（新装版）ｩ小学館 1988）. So the 未然形 of the verb stem expresses something that is not yet so, but could be, and then the -ゆ 助動詞 expresses that it is coming into being on its own. Similar for the causative, only the prefix instead expresses transitively making the verb stem come into being. それともちがいます？ -- Eiríkr Útlendi | Tala við mig 01:34, 20 January 2011 (UTC)

Your second take: you can make a causative irrealis, but a causative is not in itself irrealis just because it's a causative. That would be like saying that the attributive form is perfective just because there are perfective inflections of the attributive. — kwami (talk) 09:27, 20 January 2011 (UTC)

Okay, I'm with you now, and clearly I didn't express myself quite right earlier since I completely agree with your point here and never meant to imply that I thought otherwise.

I wouldn't say the causative is automatically irrealis by definition, but that it can be conjugated (and semantically serve) in an irrealis mode. If the verb stem is in the irrealis in order to take a CAUS or PASS suffix, then the stem is indeed in an irrealis mode. But then we have that suffix, so the mode of the larger word would depend on what the suffix is doing. I wouldn't put 食べられる or 書かせた down as irrealis simply for containing the CAUS or PASS, but I would put 食べられ or 書かせ down as irrealis if that's how it's being used (i.e. not adverbially). So when we're only talking about parts of words (i.e. 食べ・書か, or 食べられ・書かせ), which we'd have to limit ourselves to in order to even get the 未然形 anyway since it's never standalone, I do think it's appropriate to talk about the 未然形 as the irrealis, as described on the Irrealis page -- possibly pointing out that, in Japanese, the 未然形 is never standalone. -- Eiríkr Útlendi | Tala við mig 18:19, 20 January 2011 (UTC)

• The above discussion is beyond my technical competence, but I'm concerned that the article is no longer internally consistent. In the definitions, "imperfective" has been changed to "irrealis", but in the conjugation tables the formation rules still refer to "imperf.", which is now apparently undefined. I think someone should go through the whole article searching for all instances of "imperfective" or "imperf" and checking that it all still makes sense. 86.185.72.113 (talk) 14:55, 24 January 2011 (UTC)

I did that when I changed the wording. It was inconsistent before the change. The remaining instances of IPFV do not refer to 未然形 but rather contrast with the PFV in -ta. — kwami (talk) 18:19, 24 January 2011 (UTC)

Sorry, I'm confused. Originally the relevant row of the verb "stem forms" table read like this:

group/ example	1		2a	2b	sa	ka
	使・ (tsuka(w).)	書・ (kak.)	見・ (mi.)	食べ・ (tabe.)
Imperfective form (未然形 mizenkei)	使わ (tsukaw.a)¹ 使お (tsuka.o)	書か (kak.a) 書こ (kak.o)	見 (mi.)	食べ (tabe.)	さ (sa) し (shi) せ (se)	来 (ko)

A relevant example row of the conjugation table reads like this:

	formation rule	group 1	group 2a	group 2b	sa-group	ka-group
plain negative perfective	imperf. + なかった (nakatta)	書か・なかった kaka.nakatta	見・なかった mi.nakatta	食べ・なかった tabe.nakatta	し・なかった shi.nakatta	来・なかった ko.nakatta

My understanding was that "imperf. + なかった" in the second table referred to the "Imperfective form" row in the first table. Now that the row in the first table has been renamed "Irrealis form" it is not clear -- not to me, anyway -- what "imperf." refers to. 86.174.47.178 (talk) 21:55, 24 January 2011 (UTC)

I'm sorry! I completely overlooked that. I was distracted by the first column. Will do. — kwami (talk) 22:16, 24 January 2011 (UTC)

Great, thanks for fixing that. There's another aspect of this that I find rather confusing. For group 1 verbs, two "irrealis" stems are listed, e.g. 書か and 書こ. In the "formation rule" column of the conjugation table these are not differentiated, so it kind of looks as if you can use either form for any purpose, which is not the case. Could we differentiate these stems by calling them "irrealis 1" and "irrealis 2" or something? 86.174.47.178 (talk) 01:36, 25 January 2011 (UTC)

The problem there is that the form listed as "irrealis" and ending in -o is now used exclusively for the volitional -- 書こう. I'd be more in favor of splitting those out into a separate row marked volitional and including the -う on the end, perhaps explaining somewhere in the body of the text that this is an evolution from irrealis + う, where the ending -au pronunciation shifted over time into an -ou sound. -- Eiríkr Útlendi | Tala við mig 03:44, 25 January 2011 (UTC)

Historically they're the same form, and were even spelled the same through WWII. Some grammars make that distinction, but most continue with the traditional classification. Anyway, the situation is that *au is read [ou], and since WWII has been spelled that way. — kwami (talk) 09:23, 25 January 2011 (UTC)

Indeed -- I had some fun slogging through a chunk of Natsume Sōseki's Botchan, which was written before the spelling reforms and has all kinds of fun archaisms (well, not archaisms when it was written, but anyway :).

That said, since the [ou] ending is now firmly entrenched in modern Japanese, and is used exclusively for the volitional, might it not make sense to present this as a separate form? The historical derivation can certainly be described for those who might find it interesting and worthwhile, but simply presenting [o] endings as irrealis with no explanation is confusing to the learner, as evidenced by the comments above from 86.174.47.178.

In addition, many thanks for your work, Kwami! While I have the time for smaller fixes and commentary, I don't have the time (or focus) for the extensive reworking you've been doing. Cheers! -- Eiríkr Útlendi | Tala við mig 19:13, 25 January 2011 (UTC)

On the basis of the above discussion I have added a footnote to the "irrealis" item. Please make any further changes that you feel are necessary. 86.185.76.119 (talk) 21:42, 25 January 2011 (UTC)

Yes, many English-language grammars of Japanese treat it as a separate inflection for just that reason. However, since we're using Japanese names for the inflections, I don't know if that would be a good idea. AFAIK it's still just mizenkei in Japanese. — kwami (talk) 22:36, 25 January 2011 (UTC)

Reading back over the current state of the pertinent section, I think it looks good, and certainly much better than it did. Volitional is broken out in the fuller conjugation chart, which seems clear enough, while the [o] and [a] endings are both shown in the 何々形 chart above as 未然形 as appropriate, with a note to look below. Reads well to me.  :) -- Eiríkr Útlendi | Tala við mig 23:47, 25 January 2011 (UTC)

"Spelling changes" table

I find the "Spelling changes" table in the "Euphonic changes" section a little confusing. Firstly, it is not very clear that the "archaic" spellings -- some of them at least -- are still used in plenty of modern words. Presumably only in certain cases did the spelling change, whereas the presentation kind of makes it look as if all instances of the "archaic" spelling have disappeared. Also it is not clear what the footnote "usu. not reflected in spelling" is doing in a table that is titled "Spelling changes". 86.174.40.113 (talk) 14:02, 22 January 2011 (UTC)

More unexplained asterisks

In the second, third and fourth tables of the "Euphonic changes" section, there does not seem to be any explanation of what the asterisks mean. I don't know if it's supposed to be obvious, but it ain't obvious to me... 86.174.40.113 (talk) 14:14, 22 January 2011 (UTC)

.... Oh, it just occurred to me. Are the asterisks supposed to be marking disallowed/incorrect forms? 86.174.40.113 (talk) 14:20, 22 January 2011 (UTC)

Yup. — kwami (talk) 18:22, 24 January 2011 (UTC)

Thanks, I added a reminder. 86.174.47.178 (talk) 01:52, 25 January 2011 (UTC)

Poor flow

"As an abstract and rough approximation, the difference between wa and ga is a matter of focus: wa gives focus to the action of the sentence, i.e., to the verb or adjective, whereas ga gives focus to the subject of the action. However, a more useful description must proceed by enumerating uses of these particles.

"However, when first being introduced to the subject and topic markers wa and ga most are told that the difference between the two is simpler. The topic marker, wa, is used to declare or to make a statement. The subject marker, ga, is used for new information, or asking for new information."

This section flows poorly. Sorry, I do not feel very confident about trying to fix it in case I introduce the wrong meaning or emphasis. 86.174.40.113 (talk) 14:34, 22 January 2011 (UTC)

Verbal endings

"The most classic example is the classical auxiliary たり (-tari) whose forms た (-ta), て (-te), etc. are now no longer viewed as verbal endings, i.e., they can take no further affixes."

I find this a bit confusing. "Verbal endings" is exactly what I thought た (-ta) and て (-te) were. Why should the fact that they can take no further affixes mean that they are not verbal endings? 86.185.72.113 (talk) 14:46, 24 January 2011 (UTC)

Fixed. The whole article badly needs copy editing. — kwami (talk) 18:28, 24 January 2011 (UTC)

ga, ha, mo etc.

The best explanation I have run across (I think in Martin) for -ga and -ha is that the whole problem is a red herring. -ga is a case marker, like -no, -made, etc. whereas -ha is a sentence adverb, what in Greek would be called a 'particle'. In particular the natural pair to -ha is -mo; whereas -ha marks old information -mo marks new information. Here is a nice minimal triplet: watashi ha Tanaka desu 'I am a tanaka (you can see who I am and may have been wondering what I am). watashi mo Tanaka desu 'I am Tanaka (you are surprized, thinking that there were already enough around'. watashi ga Tanaka desu 'I am Tanaka (the famous one, you know who Tanaka is, and I am telling you, that is who I am). The fact that -ga is the underlying nominative marker in all cases can be seen by nominalized sentences of the type watashi-ga nonda biiru (which incidentally also show the genitive origin of -ga). In any case, this analysis is not mentioned and instead we have the hair splitting approach of 'this type of ha, that type of ha'. In any case, the whole discussion should be cited and if anyone knows of the analysis I have just presented they may be able to help me cite it too. Tibetologist (talk) 10:15, 17 October 2012 (UTC)

-wa and -ga do seem to be governed by different levels of discourse. Martin (or whoever) is probably good as an intro. But the situation is complex, and is not settled even among discourse-based linguists. — kwami (talk) 07:16, 14 December 2012 (UTC)

Contradictory right off the bat?

So... right in the first section (under "Word order"), we have the following:

"Translating the phrase the man who was walking down the street into Japanese word order would be street down walking man."

And then immediately in the next paragraph:

"The reason for this is that in Japanese, sentences always end in a verb—the only exceptions being a few particles such as ka, ne, and yo. ka turns a statement into a question, while the other sentence-final particles express the speaker's attitude towards the statement."

Aren't these in direct contradiction with eachother? If the latter is true, then the word-order from the former should end in "walking"... ne? 65.31.117.101 (talk) 05:33, 14 December 2012 (UTC)

No, the first is a noun phrase, not a clause. (It should be clauses end in a verb, not "sentences", but oh well.) — kwami (talk) 07:18, 14 December 2012 (UTC)

Er... It sounds like you're saying the same thing, then: That one of these statements is incorrect. Anyways, I don't know bupkis about Japanese grammar, so I'ma leave it for someone else to fix. :) 65.31.117.101 (talk) 06:29, 16 December 2012 (UTC)

No, they're both correct, but "sentence" is a poorly defined term and shouldn't be used if we want to be precise. — kwami (talk) 20:11, 16 December 2012 (UTC)

Sentences that don't end in a verb

The article says "... in Japanese, sentences always end in a verb—the only exceptions being a few particles such as ka, ne, and yo."

However, are there not cases in Japanese of inversion, where parts are swapped around from their normal sentence order, and hence the verb does not come at the end? Unfortunately I cannot bring to mind any examples now, but I feel that I've seen it. 86.160.83.85 (talk) 02:49, 31 August 2013 (UTC)

Yes, there's a common kinda afterthought pattern where you don't state the topic in the beginning, but then tack it after the verb. Very common with asking a question, and following the verb with its agent to clarify who you're talking about. There's commonly a drop in pitch to signal the change. Kinda like "that dog, I never could train him" -- we wouldn't say English has an OSV order exactly, but it's not true that English sentences always start with the subject.

There's also a problem with the word "sentence". A sentence is an element of written language. People don't speak in sentences. — kwami (talk) 05:40, 31 August 2013 (UTC)

The answer to your question is Yes. Japanese people do change word order. The example of the untrainable dog is a good example of why English is a word order language and Japanese is not. In the example, the order is STILL S-O-V: I-train-him. You can't get away from it! It is why English speakers feel lost without pronouns and have so much difficulty with Japanese. It is true about the "after-thought" pattern, but it's more about adding clarity than making an after-thought. I would not say there is a change in pitch. I would say there is a change in tone of voice. When you change the word order in Japanese, it usually indicates that the speaker is upset or excited, so the tone changes, and sometimes the volume too. It usually takes the form of saying the verb or predicate first, for emphasis. "おいしいいね、このごはん!" "捨てんなさい、ごみを!"

を

The article transcribes を in some cases as wo, in some cases as o. Shouldn't this be unifyed? --Jonah (talk) 11:45, 1 May 2015 (UTC)

Topic, theme, and subject: は wa and が ga

I am still in the early stages of learning Japanese. as was reading the section listed. I was a little confused by the translations since none of the sentences in Japanese seem to have a verb. One was inserted into the translation "is". Is this normal that when a verb is not present that desu is understood? — Preceding unsigned comment added by 173.166.173.129 (talk) 15:34, 18 September 2015 (UTC)