yudantaiteki Wrote:Usually transitive verbs can be converted to a corresponding passive, but you can't do that with 道を行く (i.e. you can't say *道が行かれる, or 道が行ってある). This is why JSL describes 行く as operational (i.e. volitional) but intransitive. DBJ deals with the issue by providing four definitions for を, the first of which is "marks a direct object" and the other three covering things like 道を行く.

What's the big deal? Some japanese verbs like 行く, 飛ぶ and 分かる can be used transitively or intransitively. From wikipedia:
"Verbs that are usually monovalent, like to sleep, cannot take a direct object. However, there are cases where the valency of such verbs can be expanded, for instance in He sleeps the sleep of death. This is called valency expansion."

When a verb is intrasitive, a passive construction can't be created because there is no object to become a subject. If the verb is being used transitively then a passive construction can technically be made, or at least your rule doesn't hold for all verbs. Here's an example from Japanese wikipedia (note 歩かれる):

編集] 中世. 鎌倉時代 各地に鎌倉街道が整備され、当郡を縦断し鶴川・柿生・黒川など各地を通って多摩郡へ抜け国府（府中）へ至る様々な道が歩かれるようになったと推定されている。

I'll admit that it I could only find one example for 行くthough:
せめて、A10間での道が行かれるならば行ってみたいです!

泳ぐ is also scarce but I found one:
権力を振り回した人に泳がれたら、海が汚れます!!!｣

But I expect that the reason for this is simply that you unlike the example for walk, you would almost never need to use passive constructions like "~is gone" or "~was swum" even in English. I don't see how volitionality is relevant here. It's not as if　students would be going mad writing all sorts of sentences like "the road was gone by him" or "the sky was flown by ufo's".

thora Wrote:I wanted to stress that the grammar yddt describes in this forum is accepted stuff. The concepts are not some exotic JSLisms. JSL is known to be excellent for grammar. (It just uses it's own lingo which tends to isolate it a bit.) It seems Nest0r had aready slipped in with a pile of additional general references while I wasn't looking, though.

Accepted stuff? Universally? The same 4 or 5 names keep getting mentioned. And Shibatani (in one of nestor's links) agrees with me at least on the double subject constructions, ie. a big predicate with a nested small predicate. If I recall correctly from one of the previous threads, Niwa Saburou also agrees with me. Where some of these linguists disagree with me it's because their assigning semantic meaning to words such as 'subject' and 'object' where I'm only concerned with them syntactically. I've also pointed to various examples that suggest that some of these 'rules' regarding volitionality don't accurately describe the language.

thora Wrote:Another "fact" eh? :-) A different view is that verbs have both syntactic and semantic arguments. English has different kinds of objects. You could say verbs have syntactic object arguments and the different labels for those objects are semantic distinctions. The direct object is the object argument of transitive verbs. We don't consider prepositional objects to be direct objects, for example.

I accept that the syntactic label 'object' can perhaps be broken down further into semantic categories. 'Object' in the sentences "I feel sad", "I eat an apple" and "I told him" seem to relate to the verb in different ways. But I'm yet to see an example where these semantic distinctions perform any useful function, for example helping us to understand a sentence. Even if there are some cases where it is, I'm yet to see an example where any semantic distinction of different object types breaks the syntactic ownership that the subject has over its predicate which is what this whole debate basically stemmed from (such as an を marked word becoming a subject or visa-versa).

thora Wrote:Returning to 分かる: 英語を分かる　and 英語が分かる　are both okay and mean the same thing. So is that syntactic or semantic? Is 分かる　transitive or intransitive?

Both are okay, but one is using わかる transitively, and one intransitively. The dictionary definition of わかる allows for this.

thora Wrote:Incidentally, there are some linguistic theories that avoid the "syntactic subject vs semantic object" issue by assigning case according to certain pragmatic considerations and specific ordering rules.

That sounds exactly like what I was doing with my programming analogy before. That is I looked at the ordering of an English sentence while completely ignoring semantics, and just noting that the predicate (with it's verb and object arguments) is always subordinate to the subject and allowed how this translates to meaning to depend entirely on word definition alone.

tzadeck Wrote:So the questions you have to ask are 1) Do the two systems work equally well? (This could mean 'Do students who used one system do a better job at using Japanese with fewer mistakes?') 2) Does one of the systems have more inconsistencies? 3) Is one of the systems more complex?/Does it have more exceptions?

Saying が always marks the syntactic subject and を always marks a syntactic object has less inconsistencies, is less much complex and seems to have no exceptions. Of course, I can't prove that students following this system would produce better japanese though I suspect they would because they wouldn't need to spend a whole lot of time tangled in a web of inconsistent grammar rules.

@Thora - That was funny. ;p

At any rate, it's pretty evident by nadiatims' posts that no matter what evidence or logic they're shown they won't accept it, will take and discard the most superficial aspects (ignoring resulting contradictions) of any given source, will ignore where comments clearly debunk them, will spin evidence or reword arguments, etc., anything in order to not be wrong, rather than accept they were wrong in places, overly narrow in others, and hadn't thought things through.

Case in point:

nadiatims Wrote:Accepted stuff? Universally? The same 4 or 5 names keep getting mentioned. And Shibatani (in one of nestor's links) agrees with me at least on the double subject constructions, ie. a big predicate with a nested small predicate. If I recall correctly from one of the previous threads, Niwa Saburou also agrees with me. Where some of these linguists disagree with me it's because their assigning semantic meaning to words such as 'subject' and 'object' where I'm only concerned with them syntactically. I've also pointed to various examples that suggest that some of these 'rules' regarding volitionality don't accurately describe the language.

First no one, then only 4 or 5 names? Those names get thrown around because they're established, influential Japanese linguists who have published for decades. The other half of that is people like Chino, Makino and Tsutsui, and pretty much anyone else who has written a text on Japanese grammar. They disprove everything you've said on every level from syntactic function to semantic function, and how they intersect. Shibatani doesn't “agree” with you. They say it's always intransitive. You say it's transitive. Even when you agree with what they claim is intransitive, they distinguish it as different from a subject. They point out its highly dependent nature and break it down into recognizable elements such as volition. Likewise the entire idea of transitivity as having gradations relies on volition where relevant. And it's large and small subject, in their terminology, not predicate. These “rules” accurately describe the language through and through, and what's more, they've been widely applied, of which, in English-language pedagogical texts, using the English terms like “direct object” are used in a loose, always comparative fashion to underscore the endemic Japanese functionality.

What you say is inapplicable except on the most rudimentary level and since it's just you, it really can't even be called fringe or marginal. And yet your comments have often been informally abrupt and dismissive of other posters, saying outright they're wrong and you were just schooling them, but you guess if you must you can keep at it and school them some more.

You also seem to have conflated ideas of valency and transitivity across languages and to boot you are misusing them, not understanding the underlying themes with regards to components such as volition. Those very same mechanics are described in the numerous linguistic texts to demonstrate results that are continually the opposite of what you claim—effectively even when they disagree with one another they agree where you disagree, using the methods in a rigorous, robust manner that you excised from a wiki. Not to mention, you repeatedly claim to have found exceptions to a rule that allows for exceptions, despite the bulk of your examples either being outright wrong and inapplicable to nonsensical, and always extremely rare. There are at least 440 results on Google for “swumming,” which is more than some of the examples you showed. I guess textbooks and references will have to be rewritten, despite the fact that descriptive, functional, lexicogrammatical tendencies inform these guidelines in the first place.

nadiatims Wrote:How do you know whether a verb is volitional or not without knowing its meaning?

I'm interested in this question. How do I know whether a verb is volitional or not? I go further than nadiatims by saying that knowing its meaning is not enough: ある・いる・死ぬ・喜ぶ・できる should be non-volitional judging by their meanings, but they are not, according to you guys. It seems to me that the only way to know is by checking if they have a volitional form; but that is circular.

Note that you cannot build a list of volitional/non-volitional verbs because volitionality seems to depend on the context, and unlike transitivity, changing volitionality does not change the meaning of the verb. Example: 閉まる is considered non-volitional [1], but it can be volitional sometimes, with the same meaning 「エレベーター、早く閉まれ！」

@iSoron - That link has 閉まれ crossed out, saying, “無意志動詞は《命令》をあらわす形にできない” — the example you described, shortened to “早く閉まれ,” only results in 200 hits on Google, even after including omitted results. Edit: You get the same number if you search for “閉まれ” while filtering out “閉まれば,” which is not imperative. Pardon me, that's 400-700 for the latter search.

Personally I consult grammar to establish a descriptive basis before learning, of which, rather than case markers and simplified terms, I look to explanations of dynamics and what causes elements to be ungrammatical or awkward, of which nonvolitional is a part, and retroactively I look to the same sorts of explanations when refining my understanding. Thankfully those explanations are easy to parse because linguists and teachers have chunked common patterns into categories such as nonvolitional. I use these more than I use notions of direct objects, myself.

My only interest in this topic is to defend an extra descriptive layer beyond case markers to narrow down contextual and mechanical forces in the language, and to maintain a flexible conception of the intersections of syntax and semantics bent towards functional ends so that this logic continues to be implemented in the construction of grammatical texts in the future, even as user-customized, native examples might be integrated with them in a more realtime fashion. I think in that respect many of us have similar aims.

Edit: As an aside, another link which I referenced before without the URL: Iwasaki on non-volitional/vて-いる/ある forms in terms of progressive vs. resultative/stative. You'll notice that while the most influential and salient linguists are often referenced, there are references within references representing discourse within the language literature over the decades that continually sharpens itself with exceptions and rigor based around usage and linguistic methodology, and which trickles into useful grammar texts whether the terminology is retained or not.

It's circular to use terms like direct object and particles such as を to define one another and prove each others' existence, dictating one's interpretation of a text, but it's not circular to observe and catalog language properties and use emergent terms to describe those properties where convenient and effective.

Quote:It seems to me that the only way to know is by checking if they have a volitional form; but that is circular.

It's not circular because there are other aspects to volitional verbs than just having volitional forms. Actually one of the standard ways in linguistics of testing "Is this word an X" is to see if it occurs in some patterns that other X's occur in. ("Volitional" is kind of a bad term because of confusion with the volitional form; as I've said earlier, JSL uses the term "operational verbal".)

That's really all grammatical labels are -- a collection of words that can occur in the same pattern or patterns. And nothing is absolute, there are always exceptions and gray areas. If the existence of exceptions is enough to disqualify a definition or term from being used, then grammar description is impossible. I am not saying, nor have I ever said, that you can classify every single verb as either "volitional" or "non-volitional" and the patterning of each one will be completely distinct, with no exceptions or overlap. But I find that even the exceptions can be explained in terms of the volitional/non-volitional distinction.

Quote:Example: 閉まる is considered non-volitional [1], but it can be volitional sometimes, with the same meaning 「エレベーター、早く閉まれ！」

One aspect where I think JSL's treatment is out of date is that one of the places where you see overlap between volitional and non-volitional is when it comes to mechanical (and especially automatic) things. Even though in general 閉まる is non-volitional and does not occur in volitional patterns, in the case of an elevator, the fact that it's closing automatically (and under the power of a motor) is apparently felt by (at least some) native speakers to be a kind of volition.

Although there is some context involved, in most cases no context is needed, or it requires an extremely unusual and unlikely context to turn a non-volitional verb into a volitional one.

yudantaiteki Wrote:Actually one of the standard ways in linguistics of testing "Is this word an X" is to see if it occurs in some patterns that other X's occur in.

I have no problem with that direction: check if the patterns are acceptable → conclude that the verb is volitional (in some specific context). But you will agree with me that, for learners, the most useful direction is the other one: check if the verb is volitional (in some specific context) → conclude that the patterns are acceptable; and I don't see how one can do that. As nadiatims said, that information is not included in the dictionaries; guessing from the meanings does not work; and the few short lists available are not reliable.

It's true that dictionaries typically don't list it, but that doesn't necessarily mean it's not a useful distinction. Categories can be helpful for making sense of what you learn, not just what you can explicitly get from reference works. Considering the volitional/non-volitional difference affects many things in the language, you can learn a lot about how a verb might be put to other uses by seeing whether it takes a direct object (for instance).

There are other distinctions in the language that aren't typically listed in dictionaries, like the verbs that can't take the progressive meaning with -te iru (i.e. 死ぬ or 行く), but it doesn't mean the distinctions aren't there or aren't useful.

@iSoron

Concepts like whether a verb is non-volitional don't strike me as the sort of thing you should try to make into lists to memorize and continually apply in a top-down manner, but instead as a useful concept for understanding language mechanics. I wouldn't want to use a grammar book, or would use it only marginally, if it wasn't designed like DOBJG or JMW or something like that and used such strategies. I find them very useful. That's for me. But I'm just a beginner, so I can't speak for intermediate or advanced users.

Also, all of the lists I've seen are quite reliable and make sense when I think about the word.