Please help me create Japanese _sentence_ diagrams for beginners

[Rrwrex]

@kokopelli121123

I like your translation at the top where you had the sentence more dependent on the final verb.

I’m afraid I’ve lost the fox here. Which translation at the top of what?

@kokopelli121123

I’d be curious to see if a sample of native speakers were interviewed if they viewed the 私 as a subject or a predicate.

This interests me, too. I think it’s sort of Heisenberg-grammar: the bare-word noun exists in both states at once (a subjeticate).

I think making it a rule for diagramming that a predicate must always be explicitly provided makes sense though. I think @theghostofdenzo mentioned that assuming だ is simpler with the nose-ring poem, and I agree.

Also, if we assume the provided word is the subject, then exactly what verb is correct depends entirely on context. I think this is why many objected to my がある diagram with the nose-ring senryu.

If the correct verb changes depending on context, then the same diagram won’t work for the exact same “sentence” in different contexts. The same diagram always works if we assume it’s a predicate with だ (and a zero pronoun for subject).

I’ve convinced myself that assuming だ is the correct choice.

[Rrwrex]

I’m learning important things from pretty much every single page in this wonderful wonderful middle-school grammar course that @theghostofdenzo (or was it @pm215?) pointed out.

What a fantastic resource! It’s helping me tremendously as I work through how to diagram things from a Japanese pedagogy perspective.

Today’s useful vocabulary words: 連体(れんたい)修飾語(しゅうしょくご) (modifiers that affect 体言(たいげん) or nouns/subject-things) and 連用(れんよう)修飾語 (modifiers that affect 用言(ようげん) or verbs/predicate-things).

I’m still diagramming both the same way (just modifiers hanging below what they modify) but it was especially gratifying to validate my initial instincts.

I was somewhat nervous about diagramming objects no differently than any other modifier. I was also nervous about even using the English word “modifier”! As an English speaker, it seemed pretty audacious to say that grammatical objects should be diagrammed just like adverbs, but that’s exactly how it’s taught in Japanese grammar: を marked words are 連用修飾語 (see 2.2 on that page, the example sentence 私は、種を連用修飾語 たくさん連用修飾語 まく用言 in particular）.

I’m going to end up refactoring much that I’ve already published. Great artists steal: I’m planning to pretty much mirror that website in the Grammar section. Hopefully, the diagramming section can then become much less verbose and mostly just show examples.

Onward!

[Rrwrex]

@Arzar33

I was wondering why you put the modifiers on a line going toward the bottom-right,

Holy crap!

These work SO MUCH BETTER HANGING IN THE OTHER DIRECTION!!

Sample:

The “Reed-Kellog” system blinded me to the obvious. They should definitely hang the other direction in Japanese diagrams. Far more readable.

The site is a bit of a mess at the moment (lots of broken links and some major restructuring going on) but JUST LOOK HOW MUCH BETTER MODIFIERS LOOK HANGING THE OTHER DIRECTION: Diagram basics 👈 Bunzu

[Rrwrex]

EDIT: Sorry for the rewrites – this is almost stream of consciousness. But I think I’ve got it now.

@theghostofdenzo

I can’t bring myself to think of the て form as adverbial anymore than the “and” in something like “I went to the shops and bought an apple”

@kokopelli121123

I still see て as dependent on the final verb. […]

新宿に行って、買い物して、忍者を戦って、帰ってから寝た。

I’m finally back to focusing on this.

So I’ve convinced myself that most conjugations/inflections still create a single 用言(ようげん) and shouldn’t get any special handling on a clause line. I’ve even given up on the dictionary dot thing (didn’t seem to add much value).

But the て form verbs we’ve been looking at are actually separate clauses. These are compound sentences to my way of thinking.

If I understand correctly, the て examples use something called 連用形(れんようけい) (“continuative form”). It’s discussed in section 2.2 here: 動詞の活用形とその用法をマスターしよう - 国語の文法（口語文法）

The example they give is:

彼(かれ)に 会(あ)って 真相(しんそう)を 確(たし)かめる。
I’m going to meet him and find out what’s really going on.

The て in bold indicates where the 連用形(れんようけい) inflection occurred. The plain form verb is, of course, 会(あ)う.

The word “continuative” seems to explain why we typically us the word “and” in our English translations (“go and shop and fight and then return and sleep”). It also at least suggests a continuation into (if not dependence on) the final verb when a bunch of 'em are strung together as @kokopelli121123 argued.

It also highlights the fact that Japanese sentences/clauses are quite often just single verbs (English clauses pretty much always require both a subject and a verb).

連用形(れんようけい) (continuative form) verbs can become 連用修飾語(れんようしゅうしょくご) (verb modifiers) per my dictionary:

活用形の一つ。連用修飾語となる。また、文の中止に用いられる（連用中止法）。また、助動詞「た・たい・ます・そうだ（様態）」、助詞「て・ても・たり・ながら」などを付ける。文語では、助動詞「き・けり・つ・ぬ・たり・けむ・たし」、助詞「て・つつ・ながら」などを付ける。

In other words: 会って is the “continuative form” of 会う, but it’s acting as a modifier (連用修飾語) for the verb 確かめる in this sample sentence.

It’s now much clearer to me that each of those て pieces of either example (the fight-a-ninja-and-sleep example or the talk-to-him-and-find-out-what’s-really-going-on example) are well and truly separate clauses. Those are compound sentences.

Here’s how I’d diagram the simpler sentence from the article I linked to above:

彼(かれ)に 会(あ)って 真相(しんそう)を 確(たし)かめる。
I’m going to meet him and find out what’s really going on.

The core clause is just 「確(たし)かめる」

There are two 修飾語(しゅうしょくご) (modifiers)

• 真相(しんそう) (the facts; the real situation) which is an object identified with を

• An entire separate clause which must have the predicate inflected to 連用形(れんようけい) (て form) in order to act as 修飾語. The core of this clause is just 「会って」and it has its own modifier, 「彼に」 indicating who they are meeting with.

• I thought about “sharing” the subject by extending the vertical line in the top clause down, but it’s possible to add an explicit subject to either or both clauses

  彼女が彼に会って私が真相を確かめる (She’s will meet him and I’m going to find out what’s going on)

Each clause has a grammatical subject, but since both are omitted and neither is explicitly stated, it’s safe to assume the same subject. The dotted line indicates both “@” symbols can be assumed to refer to the same person/thing.

[Arzar33]

@Rrwrex

These look SO MUCH BETTER HANGING IN THE OTHER DIRECTION!!

Great, thanks for trying!

@Rrwrex

I was somewhat nervous about diagramming objects no differently than any other modifier. I was also nervous about even using the English word “modifier”! As an English speaker, it seemed pretty audacious to say that grammatical objects should be diagrammed just like adverbs, but that’s exactly how it’s taught in Japanese grammar: を marked words are 連用修飾語

Sure ! The direct object is not that special. Personally, I even think the subject is not that special either in Japanese, it also just modify the verb and is marked by the particle が、like how the direct object modify the verb and is marked by the particle を.

But well, if I understand correctly that’s not even how 国語文法 thinks about it, they do a 主語/述語 distinction. What worry me a little when reading the 主語 wikipedia article, is that apparently the distinction 主語//述語 concept itself was basically imported from the West during Meiji era and used as a base to construct the 国語文法, so maybe it was retrofitted in a language in which it doesn’t work that well. I don’t really see why the subject couldn’t be seen as another modifier of the verb, answering the question “who did…”

For example, with a sentence like 今日私がフォークでにんじんを食べた。(probably terribly unnatural, 私が is here for exposition purpose), to me it feels like “食べた” is the core of the sentence and everything else modify it, either bound explicitly by particle 私が/フォークで/にんじんを or adverbially like 今日… (like I sketched here)

But anyway, identifying the subject is always so tricky in Japanese that maybe putting focus on it like Reed-Kellog diagram does is not a bad thing.

@kokopelli121123

新宿に行って、買い物して、忍者を戦って、帰ってから寝た。

@Rrwrex

Something like this maybe?

I’m still wondering what’s the best way to deal with implicit subject. In this example, the subject of 行って, 買い物して, 戦って, 帰って and 寝た stays the same, so it’s fine to have only one [＠が] on the baseline.

But implicit subject can change so quickly in Japanese, I feel that there is many sentence in which we can understand every word, every grammar, do a nice sentence diagram of them, and still have no clue what it’s mean because we don’t know who is doing what. Here are some sentences that gave me trouble along he year, or to other people on Wanikani.
話に乗ってくると取っていい？
トムさんはいくら注意しても、言うことを聞かないので困っています。
(context: あの女の子は指導に従うようなことはなく、好きに泳ぐし好きに歩く。) 最初は大人も注意していたけれど、今は諦めて放っておかれていた 。

@Rrwrex

(I think it usually implies a sequential ordering, but I think simultaneous events are also possible.)

I think so too, I remember seeing an example going like this 朝ご飯はパンを食べてコーヒーを飲む。Does’t mean that I eat bread first and then drink coffee, but that eating bread and drinking coffee are the things I do at breakfast, more or less at the same time.

[Rrwrex]

@Arzar33

identifying the subject is always so tricky in Japanese that maybe putting focus on it like Reed-Kellog diagram does is not a bad thing.

Yes!! Oh, ten-thousand-fold yes! This precisely scratches the itch I’ve had for so long.

My intent with these diagrams is to help English-speaking learners of the language. The constant omission of subjects is one of the weirdest parts of the language coming from an English background.

The ambiguity from this omission puzzles us constantly (as evidence, I submit the daily senryu thread in its entirety!).

@Arzar33

apparently the distinction 主語//述語 concept itself was basically imported from the West

Yes, I’ve heard (and believe) this, too. I’m not too worried about it though, because I think it’s fair to say that every comprehensible clause in any language always has a doer and an action/state, a subject and a predicate.

I think this is a logic thing more than a linguistic thing, so I’m comfortable with treating が specially.

@Arzar33

I’m still wondering what’s the best way to deal with implicit subject.

Yup. Great minds think alike. Realizing this made me re-write that entire reply.

@Arzar33

朝ご飯はパンを食べてコーヒーを飲む

Does [that] mean that I eat bread first and then drink coffee, [or that I do them] more or less at the same time?

Yes.

(Laugh)

I think that ambiguity exists in both languages (て form or “and”-connected clauses). I think it usually implies a consecutive ordering, but sometimes means simultaneous. It’s ambiguous.

[pm215]

@Rrwrex

I think it’s fair to say that every comprehensible clause in any language always has a doer and an action/state, a subject and a predicate.

I think this is a logic thing more than a linguistic thing

There’s no actual “doer” in the sentence “It’s raining” – there’s a subject “it” because English grammar demands one, but that’s more of a linguistic thing than a logic thing (Wikipedia describes this under Impersonal verb.)

Imperatives are also interesting: in “Don’t touch that!” although there clearly is a doer of the action (the person being spoken to) they don’t grammatically appear in the sentence.

[Arzar33]

@pm215

Imperatives are also interesting: in “Don’t touch that!” although there clearly is a doer of the action (the person being spoken to) they don’t grammatically appear in the sentence.

Actually I was wondering how sentence diagram work in English in that case. It looks like the traditional way is to either replace the missing subject with a X, like “X | don’t touch that” or with the understood subject in parentheses like " (you) | don’t touch that"

[Rrwrex]

@Arzar33

or with the understood subject in parentheses like " (you) | don’t touch that"

I’ve been using square brackets for implied/understood things. I’m not sure if that’s something poorly remembered from my childhood instruction on diagrams or if I stole it from quoting conventions.

@pm215

There’s no actual “doer” in the sentence “It’s raining”

I’d say the logical subject is “the weather” in that sentence. The syntactical subject is “it” which stands in for “the weather”.

If “is” is a copula, then what is it coupling if there is no subject?

Admittedly, “doer” is a poor explanation for sentences where the predicate isn’t an action, but “doing or existing in some state” gets to be wordy.

@pm215

in “Don’t touch that!” although there clearly is a doer of the action (the person being spoken to) they don’t grammatically appear in the sentence.

Yup. Clear example of an implied (or understood) subject to me. That’s how I’d diagram it. I think the core clause is “[you] touch that” and I think “do not” would somehow modify “touch” in an RK diagram, but I’m unsure exactly how it would be diagrammed.

I prefer making it a hard rule in diagrams that there always must be a subject in a clause (just as RK diagrams have the same rule). Putting implied subjects in brackets seems better than eliminating the rule.

[Rrwrex]

I just created a new diagram in the daily senryu thread.

This was my first attempt at creating a diagram for something with a 〜ば conditional, but I think it makes sense.

Diagram repeated here for convenience

趣味(しゅみ)日舞(にちぶ) 詳(くわ)しくきけば盆踊(ぼんおど)り
If you ask [@ about their] traditional-dance-hobby in detail, [it’s] bon-odori

Thoughts?

[pm215]

So, I think it ought to show the implicit subject in the conditional clause, i.e. “@が” (the obvious implied subject being 私, – “when I asked (them) about…” The “@に” is the implied indirect subject, and I thought they were shown by lines that attach diagonally?

[Rrwrex]

I struggled to parse your question at first (problem almost certainly on my end) but yeah, I think you’ve caught a mistake.

Here’s my updated understanding:

The conditional clause is 「趣味(しゅみ)日舞(にちぶ)詳しく聞けば」

• The core of that clause is just 「聞けば」

• The subject of that clause is the zero pronoun, but I’d indicated this with に rather than が. This was a brain fart on my end, but now I think that’s incorrect. It should still be ＠が as you point out.

  • If someone said 「聞けば？」it could mean “if you ask her” or “if she asks him” or whatever (possibly talking about a conversation in the third person).

  • I think my mistake was automatically thinking 「彼女に聞けば？」 or something similar, and thinking of 彼女 in that form as a subject (but it’s not). A full form with an explicit subject would be the awkward sounding 「あなたが彼女に聞けば？」

  • I think this senryu could mean “if you ask me about my traditional-dance-hobby, it’s bon-odori” but it could also mean “if you ask them about their traditional-dance-hobby, it’s bon-odori” (my brain went toward the latter).

The reply clause is just 「盆踊り」

• As discussed, this implies 「[＠が] 盆踊り [だ]」

• The syntactic subject is the zero pronoun, but the logical subject is “their traditional-dance-hobby” in this case. That’s why I’ve drawn the dotted line to 趣味日舞 and not to the subject of the conditional clause. The coupling is between what was asked and bon-odori, not between who was doing the asking and bon-odori.

Here’s the corrected diagram (with が instead of に):

---

For contrast, here’s one for a slightly different (simpler) sentence:

[Rrwrex]

Re: every clause has a subject

@pm215

There’s no actual “doer” in the sentence “It’s raining” – there’s a subject “it” because English grammar demands one, but that’s more of a linguistic thing than a logic thing (Wikipedia describes this under Impersonal verb.)

I now understand my confusion when I first read this: you had a typo. You intended to write “it rains” not “it’s raining”.

The sentence “it’s raining” absolutely has a subject (“it”).

The sentence “it rains”, though, doesn’t have any logical/semantic subject. That sentence uses an “impersonal verb” or “weather verb” per the wikipedia article (that I’m only just now reading).

I stand corrected: some sentences don’t have any logical subject (at least not in English). “It rains” is one example.

But the RK system definitely mandates a syntactical subject for every clause, and we appear to do this even in English. Despite any logical meaning, the syntactical rules force us to automatically add a “fake” subject. We say, “it rains” and not simply “rains” even though there is no real subject.

〜て form continuatives

@Arzar33

@kokopelli121123

新宿に行って、買い物して、忍者を戦って、帰ってから寝た。

@Rrwrex

Something like this maybe?

I’m still wondering what’s the best way to deal with implicit subject. In this example, the subject of 行って, 買い物して, 戦って, 帰って and 寝た stays the same, so it’s fine to have only one [＠が] on the baseline.

Okay, I think I’ve figured out how to diagram these. It mirrors the English way of diagramming conjunctives connecting clauses in compound sentences.

@kokopelli121123, @theghostofdenzo, @pm215, @Arzar33, @Escalus I’d appreciate it if you’d review and let me know your thoughts on this: The menagerie 👈 Bunzu

I think it holds together now. That example only has one continuative, but with a sentence like the 行って、買い物して、… example, you’d just stack the independent clauses the same way (and dotted line connect all the subjects).

Interestingly, it also works the same way for the conditional form connecting multiple clauses: The menagerie 👈 Bunzu

Thoughts?

[Escalus]

Continuing the discussion from the senryu thread -

Is there any situation whatsoever where you would agree that a noun is just a noun, and not part of an implied sentence?

[pm215]

@Rrwrex

The sentence “it’s raining” absolutely has a subject (“it”).

The sentence “it rains”, though, doesn’t have any logical/semantic subject. That sentence uses an “impersonal verb” or “weather verb” per the wikipedia article (that I’m only just now reading).

No, I think that all of “It is raining”, “It rains”, “It rained”, “It used to rain” etc ought to be analysed the same way. I think the simplest thing is calling the ‘it’ a dummy subject which is there for syntactic reasons, but apparently the “‘it’ refers to ‘the local weather’” theory has support with some linguists. But whichever you prefer I think the same thing applies to all of these, which are just the same verb with different tenses.

[Rrwrex]

@Escalus

Is there any situation whatsoever where you would agree that a noun is just a noun, and not part of an implied sentence?

Not if it’s intended to communicate a complete thought. At least I can’t think of any example. [Please let me know if you think you might have an example of a noun not being part of an implied sentence.]

If someone said “apple” out of the blue, in an empty, windowless room, with absolutely no other context, I wouldn’t interpret it as a sentence/clause, nor a complete thought. I’d be awfully confused, though.

If it were part of of a conversation, though, I’d interpret it differently.

In this exchange, for example:

A: Name a fruit.

B: “Apple”

I’d absolutely interpret B’s response as meaning “An apple is a fruit”.

Similarly, the senryu about a nose-ring was absolutely trying to communicate something about the object. I think it has to be interpreted as a sentence to communicate any sort of thought.

Even if the thought is simply “being”:

Cogito, ergo sum
“I think, therefore I am.”

“Sum” (“I am”) is either a profound thought or a simple one, but it’s a complete clause with an implied subject and an explicit verb.

(Not a great example since that’s a verb rather than a noun, but I find it an interesting example regardless. Apparently Latin, like Japanese, also uses single-word sentences (albeit with verbs) to express complete thoughts with implied subjects.)

[Rrwrex]

@pm215

No, I think that all of “It is raining”, “It rains”, “It rained”, “It used to rain” etc ought to be analysed the same way. I think the simplest thing is calling the ‘it’ a dummy subject which is there for syntactic reasons

Yeah, I think I sorta get it now.

The question is, “What is ‘it’ in those sentences?”.

Clearly there’s a syntactic subject in all of them, but is there a logical/semantic subject?

I’m not 100% certain I’d answer “no”, however. Even the wikipedia article says “impersonal” verbs are sometimes called “weather” verbs. Isn’t the logical subject “the weather” in all those forms?

[pm215]

@Rrwrex

Similarly, the senryu about a nose-ring was absolutely trying to communicate something about the object. I think it has to be interpreted as a sentence to communicate any sort of thought.

That’s an interesting point of view. Personally I think you can also communicate perfectly well with a mere sentence fragment, and I think that in that situation the speaker is not taking a complete sentence and chopping a part out to say, and the listener is not mentally reconstituting a full sentence from what they hear.

I think also there are a few things here in danger of being confused, which I would separate out as (and state my personal view as):

1. Japanese as a language is happy with sentences that omit parts that English requires to be stated. The sentences are still complete as Japanese sentences even if those parts are omitted, even if a literal English translation would result in a non-complete sentence.

2. Poetry is special, and absolutely will play around with fragments of sentences, flipping parts of sentences around, words on their own, etc, in pursuit of art.

3. Ordinary humans thinking and communicating with each other don’t always use complete sentences.

[Rrwrex]

@pm215

1. Japanese as a language is happy with sentences that omit parts that English requires to be stated. The sentences are still complete as Japanese sentences even if those parts are omitted, even if a literal English translation would result in a non-complete sentence.

2. Poetry is special, and absolutely will play around with fragments of sentences, flipping parts of sentences around, words on their own, etc, in pursuit of art.

3. Ordinary humans thinking and communicating with each other don’t always use complete sentences.

Perhaps surprisingly, I agree wholeheartedly with all three.

It’s just that if I want to explain or display how I interpret sentence fragments, I have to add the additional grammatical pieces.

The entire point of diagramming sentences is to explain how one parses something to extract meaning. If all you have is a noun then no thought is being communicated.

Calling “nose-ring” a “noun phrase” and then calling it a day explains exactly nothing. It just gives it a different name.

Diagramming it as “[it is a] nose-ring” explains how this reader interprets what the author was trying to communicate. Another reader might diagram it as “[a] nose-ring [is a work of art]” or whatever, but without expressly describing what they feel is implied nothing gets communicated.

---

This is exactly what I was trying to express in this note after the diagram for the 盆踊り senryu:

Poems aren’t necessarily complete or grammatical. This diagram shows one possible interpretation, but other interpretations are quite possible and equally valid.

A direct transliteration of all three stanzas, without attempting to fill in missing grammatical structure would just be something like “Japanese-traditional-dance-hobby / if-ask in-detail / bon-odori”.

We’ve chosen to “fill in the blanks” by interpreting this as meaning: “If you ask about the traditional-dance-hobby, it’s bon-odori”.

---

@pm215

Personally I think you can also communicate perfectly well with a mere sentence fragment

I’d like to challenge this.

“Apple” is a sentence fragment. What does it communicate?

Can you communicate something to me (anything at all) using only a sentence fragment that doesn’t clearly imply the remainder of a sentence?

(Expletives are perfectly acceptable! I’d interpret a one-word response of “Jerk!” as “[You are a] jerk”, for example. ).

[Escalus]

@Rrwrex

It’s just that if I want to explain or display how I interpret sentence fragments, I have to add the additional grammatical pieces.

The entire point of diagramming sentences is to explain how one parses something to extract meaning. If all you have is a noun then no thought is being communicated.

Calling “nose-ring” a “noun phrase” and then calling it a day explains exactly nothing. It just gives it a different name.

Diagramming it as “[it is a] nose-ring” explains how this reader interprets what the author was trying to communicate. Another reader might diagram it as “[a] nose-ring [is a work of art]” or whatever, but without expressly describing what they feel is implied nothing gets communicated.

The nose-ring is a thing. The author wants the reader to think about the thing, and the feelings associated with the thing. But they aren’t making any actual statements about the thing. “Nose-ring is” (in the だ sense of “is”) requires another thing for the nose-ring to be equivalent to… but there isn’t one. It isn’t implied by context, because there isn’t any.

“I want you to think about the phrase ‘a nose-ring’.” is a complete thought, but the phrase itself is not.