pm215 Wrote:(1) 鼻が長い isn't connecting to a particle or noun, but it's not adverbial either

This whole embedded sentence functions as an adjective. The です of the outer sentence connects it to 象が, but even without です an i-adjective can connect to a noun after it or subject before it on its own.

pm215 Wrote:(2) shuffling produces obvious nonsense: 鼻が長い象がです

この部屋が彼が狭いと言う is also obvious nonsense. That does not mean that この部屋が狭い is not an embedded sentence. You can however do 鼻が長い象が大きいです。The long-nosed elephant is big. If you move the adjective from the slot that binds to the subject this slot becomes free and must be filled again with a verb to create a valid sentence, e.g. 大きい, 大きいです or 寝る.

pm215 Wrote:(3) inflection affects the 'embedded sentence': 象が鼻が長かったです

What about 彼がこの部屋が狭かったと言う? This infection criterion seems pretty random to me.

Why not just check if the embedded sentence is really a sentence that could stand alone? Both 鼻が長い and この部屋が狭い look like valid Japanese sentences to me.

pm215 Wrote:...so the obvious conclusion would be that it isn't in fact an embedded sentence at all.

This is not obvious for me.

xaarg Wrote:Why not just check if the embedded sentence is really a sentence that could stand alone? Both 鼻が長い and この部屋が狭い look like valid Japanese sentences to me.

Because that just tells you that the fragment would happen to be a sentence if you looked at it on its own, not that it's actually functioning as an embedded sentence in the context of a larger one.

Random rebuttals to your more detailed points: (1) "the whole embedded sentence functions as an adjective" -- an adjective to what? there's no noun after it.
(2) the idea is to reshuffle the components of the outside sentence. Trying to move part of the outer sentence into the inner sentence doesn't work, as you notice.
(3) The point is that it's *possible* to put the outer verb in the past tense without affecting the inner one, not that the inner one is immune to all change. In contrast, you can't inflect the 'outer verb' of 象が鼻が長いです without affecting the 'inner verb', which suggests that there isn't in fact an inner and an outer here.

pm215 Wrote:Random rebuttals to your more detailed points: (1) "the whole embedded sentence functions as an adjective" -- an adjective to what? there's no noun after it.

Adjectives of this kind turn into verbs if they are put at the end of the sentence just like regular i-adjectives do. In English adjectives can only modify nouns that follow after it, but Japanese i-adjectives are more powerful.

pm215 Wrote:In contrast, you can't inflect the 'outer verb' of 象が鼻が長いです without affecting the 'inner verb', which suggests that there isn't in fact an inner and an outer here.

Well, there is an outer and an inner verb.

The outer verb is 鼻が長いです.

This the polite version of the verb 鼻が長い.

鼻が長い is a verb created from the i-adjective 鼻が長い. Its form has not changed, but i-adjectives at the end of a sentence automatically turn into verbs.

鼻が長い is not a pure i-adjective that you can find in a dictionary, but an i-adjective created from an embedded sentence. That sentence is composed of the subject 鼻 marked by the particle が and the proper i-adjective 長い, that acts as the verb of the embedded sentence by being sentence-final. This 長い is therefore the inner verb.

Here the inner verb is part of the outer verb, so if you change one you also change the other. The outer verb is still an (i-adjective that is [an embedded sentence that contains a {verb that is an i-adjective}]). Where does your rule that those things need to be totally independent come from?

*edit: didn't see xaarg's response which may well make better sense than mine.

I'll add brackets to your first example sentence to clearly show the embedded sentence:
[彼が[この部屋が狭い]と言う],
when you shuffled it around, the embedded sentence remained intact, and no ambiguity was introduced:
[[この部屋が狭い]と彼が言う]

in your second example:
[象が[鼻が長い]です]
you reshuffled it as:
[[鼻が長い]象がです]

The problem is that this introduces ambiguity in that the phrase 「鼻が長い」 is now acting as an adjective. Now as someone (maybe yudantaiteki?) pointed out earlier, the verb (to be) is implicit in the i-adjective. So 鼻 is still the 'owner'(abject) of the 'implicit verb' contained in the adjective 長い. And 象 is still the 'owner' of the copula です (to be). But to be what exactly? that slot has now been left empty. Unlike other elements of a japanese sentence, you can't move the contents of that slot around because at some point in time it became bound to the copula by sentence position(as in english) rather than by a particle. So we read it as ungrammatical because in the absence of context, it makes no sense. There is missing information. If however, you move the copula with the adjective, and add a comma or period to prevent the '鼻が長いです' from acting as an adjective, it starts to make sense again:
[[鼻が長い]です。象が]
or
[[鼻が長い]です、象が]
Although when not at the end of a sentence, です should become である:
[[鼻が長い]である、象が]
So 象が is acting like an after thought to fill the missing abject slot of the copula.
Now the only slightly mysterious part to me is that the copula is perhaps optional(?) because of the usual rule that i-adjectives don't require it. Note that a google search for 長いです and 長いである both return plently of hits.
Now if we remove the です part 象 becomes the abject of a missing (albeit obvious) verb, which leads me to theorize (bear with me) that perhaps there is some kind of null verb implicit in the particle が, which for the time being I'll assign the meaning 'is'. Note the only time this 'is' definition is potentially useful is in examples like the above, or when が is used with a na-adjective as part of an adjectival phrase:
顔が奇麗な人 = face (is) beautiful person = person with beautiful face
When another verb is present, the hyperthetical が null verb can be ignored.
蛇が鼠を食べた。= snake (is) ate mouse. (which sounds odd in english but who cares. I prefer translating to a kind of pseudo english that stays true to the mechanics of the japanese language.)

Any more questions?

IceCream Wrote:but what about the 好き？ am i missing something entirely obvious?

Do you mean why が is attached to 好き？

One thing that helps a lot to understand が is to understand it's history, and its use in classical Japanese. が has an association of attachment and intimacy to the subject, which is why you use it for emotions like 好き, 嫌い, etc. Those are your personal feelings, attached to you as a human being, and so you use が. It's the same idea for body parts, etc. which is why が is used in that sentence 像は鼻が長い.

Oh, maybe I should elaborate a little on what I meant by 'classical Japanese'. が used to serve a function similar to の, in that it created a connection, possession, attachment. Some examples of this still survive in Modern Japanese. 吾が for example, 我国、我家、etc, where が all means "MY" and shows possession and attachment. This is also why 我 can take the reading が

So in 私は犬が好き for example, the が shows that it's MY feeling, that I possess it, that it is intimately related to me.

Aijin Wrote:One thing that helps a lot to understand が is to understand it's history, and its use in classical Japanese. が has an association of attachment and intimacy to the subject, which is why you use it for emotions like 好き, 嫌い, etc. Those are your personal feelings, attached to you as a human being, and so you use が. It's the same idea for body parts, etc. which is why が is used in that sentence 像は鼻が長い.

I don't know about that; you can use は instead for all those things instead, and が seems to be used in many cases where there is no attachment or intimacy. 今日は試験がある, for instance, especially if you're not the one taking the test.

I was specifically answering her question regarding its use for things such as 好き. Emotions, ability, such as 出来る etc, use が because of its attachment connotations. Historically が was used to convey meanings that in Modern Japanese we have different particles for, and that myriad nature is still present in Modern Japanese. There are too many uses for が to pigeon-hole it with any single definition, and the logic and rules behind its use consequently vary between situations.

I think it's a common trap that foreigners fall into, that they try to categorize or manifest a single definition in their mind for a concept in a foreign language. For many particles, their uses are too diverse and subjective to specific situations to make universal statements about them. The rules that govern their use are going to vary between the context, and you simply have to memorize and become familiar with all the various contexts. This is the way that children learn their mother tongue, and I don't think there's any way around it. You can try to come up with a logical definition and rules for が, but ihey're going to always fall short. You simply have to experience all the uses through intense immersion, and eventually that subconcious knowledge comes.

Now, I DO think that's important for beginners to have some simple working definition for these things, to give them some ground to stand on, so to speak, but after you have a concept of が, I think it's best to simply let yourself become fluent in it through immersion.

Yeah but people get curious, and it's only natural for them to try dig up things to justify why something is something.

That was my trap with learning kanji, I dived right into physical etymology and it got to a point where I was wasting far too much time learning etymology than the characters themselves.

He who pursues knowledge in his lifetime is like that who drinks sea water, the more he drinks the thirstier he gets Ibn Al Muqafaa

Ahhhh, life is beautiful~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

IceCream Wrote:id say that this is more of a semantic problem than a logical one. Seems to me that there's no reason the (logical) grammar for が also should fulfill semantic issues like that one.

OK, but even if you get rid of that, there's still the 自分 test. 自分 can only refer to the subject, not any が marked noun. The following example is from Natsuko Tsujimura's book "An Introduction to Japanese Linguistics:

太郎が花子が自分のグループで一番好きだ。 In this sentence, 自分 can only refer to 太郎 and not 花子, showing that there is a grammatical property that 太郎 holds in this sentence that 花子 does not, despite the fact that they are both marked by が. This is one area in which the concept of the subject is important, because if we say "自分 can only refer to the subject of the sentence", we have to know what the subject is. Tsujimura shows also how 自分 can refer to は or に marked nouns, and can even refer to multiple nouns in cases where there are multiple subjects due to embedded sentences (i.e. 太郎が花子に次郎が自分を批判したと言った; here　自分 can be either 太郎 or 次郎.) This is probably the clearest way to show that there is a concept of "subject" in Japanese that transcends any particles that the words happen to be marked with, and in the end the notion of "subject" is not really very different from English.

liosama Wrote:He who pursues knowledge in his lifetime is like that who drinks sea water, the more he drinks the thirstier he gets Ibn Al Muqafaa

liosama, you're quote happy these days - dropping them in almost every post. Here's one you might like:

"I know that most men, including those at ease with problems of the greatest complexity, can seldom accept even the simplest and most obvious truth if it be such as would oblige them to admit the falsity of conclusions which they have delighted in explaining to colleagues, which they have proudly taught to others, and which they have woven, thread by thread, into the fabric of their lives." -- Leo Tolstoy

Thora Wrote:

liosama Wrote:He who pursues knowledge in his lifetime is like that who drinks sea water, the more he drinks the thirstier he gets Ibn Al Muqafaa

liosama, you're quote happy these days - dropping them in almost every post. Here's one you might like:

"I know that most men, including those at ease with problems of the greatest complexity, can seldom accept even the simplest and most obvious truth if it be such as would oblige them to admit the falsity of conclusions which they have delighted in explaining to colleagues, which they have proudly taught to others, and which they have woven, thread by thread, into the fabric of their lives." -- Leo Tolstoy

Waaa <3 Tolstoy

Here's one from another beloved author~

In every bit of honest writing in the world there is a base theme. Try to understand men, if you understand each other you will be kind to each other. Knowing a man well never leads to hate and nearly always leads to love. There are shorter means, many of them. There is writing promoting social change, writing punishing injustice, writing in celebration of heroism, but always that base theme. Try to understand each other.

–- John Steinbeck

^_^

The more I learn, the less I know.

IceCream Wrote:

pm215 Wrote:So I think he might say that the second part of "が defines the subject and は never does" was true but not the first.

Actually, it seems neither of those two things seem true. Looking at the 自分 stuff, if i've understood Yudantaiteki correctly that 自分 refers to the subject, then:
メアリーは自分で何でもします。

In all of these, and tons more, the subject is marked by は...

The way I look at that kind of sentence (and which I think Shibatani also does) is that は marks the topic of the sentence, and there is an implicit 'empty slot marked by が' which is bound to the topic. So は isn't marking the subject; the subject is left unstated but implicitly equal to the thing marked with は. (That's a bit of a subtle difference but personally I think it's a helpful distinction. As it happens I think it's helpful to second language learners, which isn't necessarily true of some of the rest of this linguistics discussion. I found it helpful, at least.)

Quote:I guess what i'm trying to say is, what makes anyone think that there can be rules regulating what the subject is in a japanese sentence?

My take on this is that there seems to be something in Japanese grammar which acts consistently enough to be worth labelling. The question is whether it's close enough to the English 'subject' to use the same label or not...

Quote:since the subject clearly isn't defined by が, but also, if we say が defines something other than the subject (the abject) we can explain が and not run into too many serious problems...

Does defining a term for 'thing marked by が' get you any further forward than the circular "things marked by が are marked by が" ?

I'll have to see if I can grab that Tsujimura book from the library.

IceCream Wrote:EDIT: in fact, i finally bothered to read the wiki entry on "subject", and found it's related to the predicate in english. So, now i'm totally confused as to a.) why we are transposing english grammar onto japanese in the first place, and b.) why は wouldn't mark the subject in many instances. e.g:
Roses are red. Subject = Roses. Predicate = are red.
バラは赤い　seems totally intuitive.
And, saying it doesn't because は marks the topic isn't too helpful either...
I guess what i'm trying to say is, what makes anyone think that there can be rules regulating what the subject is in a japanese sentence?

This.
Perhaps "subject" isn't something that makes sense in a Japanese grammar.
Take for example (and I know, it's not a natural language, but still) Lojban grammar. This is a grammar in which the concept of "subject" makes no sense at all. (Yes, perhaps a grammar of lojban with a subject could be written, but it seems to me it wouldn't be a very good one).
When the structure of the sentence changes radically, often the same division of the parts of the sentence can't be applied. Maybe English is too different from Japanese for the same concepts to apply.

dbh2ppa Wrote:Perhaps "subject" isn't something that makes sense in a Japanese grammar.

The definition of "subject" is pretty abstract. It's a noun phrase somehow related to the predicate. The exact definition depends on the language.

Japanese does have predicates. Japanese has nouns. And sometimes these nouns relate to the predicate. So it seems kinda obvious that Japanese does have subjects.

Even if you somehow reject the name, you still need to name the parts of a Japanese sentence to talk about them. "Subject" is the established name and it works even better on languages other than English like Latin, where the subject also can be omitted. Hell, there is even something you could call "the topic of a sentence" in Latin.

mentat_kgs Wrote:The more I learn, the less I know.

That's a good sign! Only in our imaginations are there no contradictions.

But it reminds of that idea that as the modern world requires us to specialize in increasingly narrow fields...needing to know more and more about less and less...until eventually one knows everything about nothing. =]

pm215 Wrote:Does defining a term for 'thing marked by が' get you any further forward than the circular "things marked by が are marked by が" ?

yeah, because as I explained countless times, it's the が marked verb that takes 'ownership' of the verb or copula. This is meaningful information that allow us to determine who is doing/being what in the sentence:

If the verb is active, the が marked word is doing the verbing.
If the verb is passive, the が marked word is being verbed.
If the clause ends in a copula (です) or implicit copula (i-adjectives) it is the が marked word that is being modified.
If no が marked word is present, then the verb/copula 'owner' is obvious from context, and in many cases coincides with the は marked word.
If a copula appears to be missing, it can be explained as either being obvious from context or being implitly contained in the が marker itself. In any case, it seems the japanese language doesn't like doubling up of the copula. As in the following example:
僕が林檎が好きです。=
[僕が[林檎が好きです]]。 僕 is missing a verb.
or [僕が[林檎が好き]です]。林檎 is missing a verb.

I thought that this 'ownership' of the verb or copula is what was meant by the grammar term 'subject', but have renamed it 'abject' (IceCream's term) because people seem to have other ideas about what 'subject' should be. Could someone give me a definition of 'subject' and tell me why it's even useful to know (even in English).

I'm starting to feel like a broken record.

yudataiteki Wrote:太郎が花子が自分のグループで一番好きだ。 In this sentence, 自分 can only refer to 太郎 and not 花子, showing that there is a grammatical property that 太郎 holds in this sentence that 花子 does not, despite the fact that they are both marked by が.

太郎 is being modified by the phrase 花子が自分のグループで一番好きだ. Because what Taro likes is a matter or his own volition, it makes sense to use 自分の to describe "his own group".
"Taro likes Hanako most in her own group." doesn't make sense because Hanako is just the passive object of Taro's liking. We would just say "Taro likes Hanako most in her group."

yudantaiteki Wrote:Tsujimura shows also how 自分 can refer to は or に marked nouns, and can even refer to multiple nouns in cases where there are multiple subjects due to embedded sentences (i.e. 太郎が花子に次郎が自分を批判したと言った; here　自分 can be either 太郎 or 次郎.)

自分を (自分 plus object marker を) can mean herself/himself/myself/itself. Note 自分 is filling the object(を) slot. Therefore in this case 自分 can't refer to 花子, because 花子 is the target of a verb rather than the 'abject'. We wouldn't say "Taro said to Hanako that jiro criticised herself." we would use the word "her" instead of "herself" because Hanako is not performing the action.
So who 自分 refers in any given sentence is either obvious from a logical standpoint or it is inherantly ambiguous. I guess you could make a point of finding the 自分 target and then call it the 'subject', but I don't really see the point. If the only way to find the 'subject' is through this 自分 test, then finding the 'subject' doesn't help us find the 自分 target and so it's not really useful for anything.


By the way, I'm quite ready to declare at this point, that I basically feel zero confusion about this topic, until someone gives me a example sentence that proves otherwise.

Has it ever occurred to anyone that one of the reasons native speakers can speak so fluently is because they aren't trying to conform to overly complicated and entirely imagined grammar rules?

nadiatims Wrote:

pm215 Wrote:Does defining a term for 'thing marked by が' get you any further forward than the circular "things marked by が are marked by が" ?

yeah, because as I explained countless times, it's the が marked verb that takes 'ownership' of the verb or copula.

(snip)

I thought that this 'ownership' of the verb or copula is what was meant by the grammar term 'subject', but have renamed it 'abject' (IceCream's term) because people seem to have other ideas about what 'subject' should be.

Well, that's the definition I was using for 'abject', except that I reckon that it's much simpler to think of 太郎が納豆が好きだ as being a sentence with one predicate/verb/copula (好き) "owned by" 太郎 and where the predicate takes a second が-marked noun. So I agree that 'subject' roughly means 'ownership', I just disagree that abject or subject or whatever you want to call the intuitive term is always marked が. Which is roughly what I was trying to say here.

Quote:I'm starting to feel like a broken record.

Me too :-)

Quote:Has it ever occurred to anyone that one of the reasons native speakers can speak so fluently is because they aren't trying to conform to overly complicated and entirely imagined grammar rules?

I don't think anybody's trying to put this forth as (a) something native speakers consciously think about or (b) something second language learners ought to try to memorise and use.
On the other hand I completely disagree that grammar is imaginary...

IceCream Wrote:

pm215 Wrote:So は isn't marking the subject; the subject is left unstated but implicitly equal to the thing marked with は. (That's a bit of a subtle difference but personally I think it's a helpful distinction. As it happens I think it's helpful to second language learners, which isn't necessarily true of some of the rest of this linguistics discussion. I found it helpful, at least.)

hmm. Actually i think it's entirely unhelpful to learners.

Maybe so. I think that as you're learning, you read a load of Japanese, you get some explanations from people/textbooks/wherever, you try to work it all out in your mind. At some point you encounter an idea that just clicks, and you say 'yes, that's exactly right'. But the click is as much a property of the pile of thoughts in your mind waiting to crystallise as it is of the external idea, I suppose.

PS: I had a look at that book by Tsujimura, and it appears to say nothing about は at all, which is an interesting omission...

nadiatims Wrote:Has it ever occurred to anyone that one of the reasons native speakers can speak so fluently is because they aren't trying to conform to overly complicated and entirely imagined grammar rules?

It has occurred to many people, but it's a myth. Native speakers of a language have rules for grammar in their brain, and some of them are very complicated. The purpose of linguistics is to figure out what those rules are, and then "pedagogical syntax" attempts to take those complicated rules and distill them into something that learners can make use of. But the fact that a stated grammar rule is complicated does not mean that it's wrong, however unsuitable it may be for learners.

Of course our rules will never be 100% the same as the ones in a native speaker's brain because they're innate and can't be explained easily. It's sort of like asking someone the exact sequences of muscle movements necessary to take a step; most people can't tell you that even though they're able to walk.

yudantaiteki Wrote:

nadiatims Wrote:Has it ever occurred to anyone that one of the reasons native speakers can speak so fluently is because they aren't trying to conform to overly complicated and entirely imagined grammar rules?

It has occurred to many people, but it's a myth. Native speakers of a language have rules for grammar in their brain, and some of them are very complicated. The purpose of linguistics is to figure out what those rules are, and then "pedagogical syntax" attempts to take those complicated rules and distill them into something that learners can make use of. But the fact that a stated grammar rule is complicated does not mean that it's wrong, however unsuitable it may be for learners.

Of course our rules will never be 100% the same as the ones in a native speaker's brain because they're innate and can't be explained easily. It's sort of like asking someone the exact sequences of muscle movements necessary to take a step; most people can't tell you that even though they're able to walk.

Could you clarify what you mean by 'innate'? I also feel that your definition of linguistics is a bit narrow (in the context of 'innate brain rules of natives') and doesn't take into account prescriptive vs. descriptive grammar.

ruiner Wrote:Could you clarify what you mean by 'innate'?

If you are speaking your native language, you are following some grammatical rules that are in your brain. They may not match 100% with prescriptive grammar, and you probably can't explain the rules, but they are there.

Quote:I also feel that your definition of linguistics is a bit narrow (in the context of 'innate brain rules of natives') and doesn't take into account prescriptive vs. descriptive grammar.

Sorry, I should have said "one purpose of linguistics".

Basically all I was trying to say is that there's a common belief that because native speakers of a language can't explain the grammatical rules of their own language (without training), that the rules don't exist. Or, that learners should avoid grammar rules.

yudantaiteki Wrote:

ruiner Wrote:Could you clarify what you mean by 'innate'?

If you are speaking your native language, you are following some grammatical rules that are in your brain. They may not match 100% with prescriptive grammar, and you probably can't explain the rules, but they are there.

Quote:I also feel that your definition of linguistics is a bit narrow (in the context of 'innate brain rules of natives') and doesn't take into account prescriptive vs. descriptive grammar.

Sorry, I should have said "one purpose of linguistics".

Basically all I was trying to say is that there's a common belief that because native speakers of a language can't explain the grammatical rules of their own language (without training), that the rules don't exist. Or, that learners should avoid grammar rules.

I'm with you on that last part (Magamo excluded ;p). For me, I simply focus on the idea that it's possible, in the context of most learning/acquisition, to view the grammar in the brain purely as learned and mutable amidst communication with the language, and thus try to emphasize finding the ways to analyze and self/other-teach it [Edit: To teach it as it's *used*]. Being (since my life on this forum) of the Gabrielatos persuasion (http://www.gabrielatos.com/), I like the idea of using corpora and technology to empower users to develop awareness and intuition of the language in a way that's balanced with materials curated by linguists (there's always a place for rigorous scientific discourse to function when needed as pseudo-originary intersubjective authority).