Diagram basics

Diagrams are intentionally very simple. You don’t need any fancy software to create them. The intent is to be able to create them with nothing more than a pencil and paper (or chalk and chalkboard).

Write the original sentence

The first step couldn’t be easier: Simply write the original sentence you plan to diagram at the top of the page.

Optionally, you can also write an English transliteration for the sentence immediately below the original.

Draw lines for clauses

Every sentence (in any language) has at least one clause (compound sentences might have multiple).

Every 文節(ぶんせつ) (clause) has a 主語(しゅご) (subject) and a 述語(じゅつご) (predicate). There can be many words (including complete clauses/phrases) that modify each of these, but at the core of the sentence there will always be a 体言(たいげん) acting as the subject, and a 用言(ようげん) acting as the predicate.

First draw a horizontal line. This will act as a baseline for the subject and predicate to rest on.

Then draw a vertical line to separate the subject from the predicate. This line should extend above and below the baseline, and should be positioned somewhere central.

Then write the core 主語 to the left of the separator, and the core 述語 to the right.

Let’s start with an extremely simple example:

花(はな)が咲(さ)く

[The] flower blooms

In English, this means “The flower blooms”. English wants the definite article “the” but there is no equivalent word in the original sentence, so we put it in square brackets to indicate that it was added.

There’s obviously only one clause, with single words for each part of that clause:

• The 主語 is the single word 花 (identified by the particle が). 花 is the master doing the action.

• The 述語 is the single verb 咲く. The action is “blooms”.

There are a few things to notice:

• If you ignore any lines or symbols and simply read what’s on the baseline, you get the core clause 「花が咲く」. (The core clause happens to be the entire sentence in this example, but that obviously won’t always be the case!)

• The particle が is explicitly included with the 主語 since it’s literally present in the original sentence (が identifies subjects).

Other 用言(ようげん)

The 述語 (predicate) won’t always be a 動詞 (verb). It could also be 形容詞 (is-verb) or 形容動詞 (だ is-verb). Consider, for example, the sentences:

私(わたし)が眠(ねむ)い

I am-sleepy

彼(かれ)が静(しず)かだ

He is-quiet

Note in particular that the predicate in that last example is 静かだ and not just だ. English speakers might want to think of だ as a single-word verb like the English word “is”, but だ on its own is not a “freestanding” word. Neither is 静か. The 用言 is 静かだ.

Transliteration not translation

The English versions of the last two examples expressly transliterated the 用言 into hyphenated “is-verbs” (“am-sleepy” and “is-quiet”). This is to emphasize the fact that these are individual 用言 and not two separate words.

“Am-sleepy” in the first example corresponds exactly to the single word 眠い.

Similarly, 静かだ becomes the single word “is-quiet”.

We attempt to show word-for-word transliterations rather than semantic translations. English often requires us to reorder or add words to keep things understandable, but transliterating this way reenforces the idea that these aren’t adjectives, they are full predicates all on their own.

Polite forms

So far we’ve used plain forms of the verbs. Using polite forms doesn’t change much:

花が咲きます

[The] flower blooms (polite)

私が眠いです

I am-sleepy (polite)

彼が静かです

He is-quiet(polite)

These are polite forms of 用言, but nothing has changed structurally.

Implied parts

Japanese sentences can leave out quite a bit and still remain fully grammatical sentences. The entire subject or part of the predicate can be omitted and the sentence will still make sense.

Implied subjects

If all that’s given is a 用言, then the subject is implied. As discussed in the grammar section, it’s helpful to think of the zero pronoun ”@” standing in when the subject is omitted.

Say, for example, the sentence was simply 「読む」(a verb meaning “read”). It isn’t explicit from the sentence who or what did the reading, so we use the zero pronoun to stand in for “he” or “she” or “it”:

読む

[@] reads

Notes:

• We use ”@” to represent the zero pronoun. We use brackets to show that neither the @ nor が are literally present in the original sentence.

• The core clause along the baseline is still just the single word read 「読む」 — the stuff in brackets is ignored. [Unless we are talking about the diagram rather than about the original sentence. In that case, it’s helpful to say「『まるが』読む」 with “air quotes” around the first bit.]

• This sentence does not mean “I read”, nor does it mean “she reads”, nor “it reads”. Without additional context we simply don’t know who or what’s doing the reading, so we use “@” to fill in. Note that using the zero pronoun this way ensures the diagram remains unchanged no matter the context the sentence was used in.

• As always, a clause must have both a subject and a predicate, even if the subject is implied and not explicitly present.

Bare 体言 (implied だ)

As we also discussed previously, one can create single-word sentences out of 体言. Doing so implies a だ at the end.

For example:

牛

[it] is-cow

Notes:

• Despite all the extra stuff in brackets, this is still a one-word sentences. We put things in brackets expressly to make the implied stuff visible. Structurally speaking, these implied parts are logically part of the sentences, but they obviously aren’t explicitly present.

• We still skip over everything in brackets when reading the clause along the baseline. This clause reads just 「牛」.

• Single words, no matter what part of speech, must be interpreted as 用言 — 牛 is part of the predicate, not the subject. This can seem odd to English speakers.

• The zero object is the implied subject. There is no way to know from the original sentence alone if the subject is “I”, “he”, “she”, or “it”. The semantic meaning comes from context.

修飾語 (Modifiers)

Japanese has a wonderful system of 助詞 (particles) that explicitly inform us of what different parts of a sentence are doing. They tell us if something is specifying a location, a starting point, an ending point, a means of doing something, etc.

The Japanese word for such modifiers is 修飾語(しゅうしょくご). They can modify any of the eight parts of speech (including other modifiers).

The thing being modified is called 被修飾語(ひしゅうしょくご).

These modifiers don’t actually change the meaning of the core clause, however. They just provide more, finer-grained information.

We “hang” modifiers below the things they modify.

Object modifiers

One particularly interesting modifier is identified by the particle を. It introduces a grammatical object, something the 述語 is acting on or with.

None of the example sentences so far have involved an object, so let’s add an one:

本(ほん)を読(よ)む

[@] reads a book

Notes:

• The original sentence in it’s entirety is 「本を読む」, but the core clause is just 「読む」. It’s the same structure we looked at earlier, but “book” just adds more information, it tells as what is being read.

• An English translation of that core clause is impossible without more context. We still don’t know who or what read the book. We don’t know whether to use a subject of “I”, “he”, “she”, or “it”. So we use the magic zero pronoun @ instead.

• English sentence diagrams have a special notation for objects, but in Japanese grammatical objects identified with を are just a modifier like any other modifier (as we’ll see). We dangle modifiers below whatever they modify/qualify.

• Unlike the stuff on the horizontal line (the core of the clause) we don’t include the particle with the word 本. Instead, we write the particle on the stem of the “hanger”. This informs us what type of modifying/qualifying is going on (sushi is the grammatical object because of the を).

• The “hangers” might look backwards to anyone familiar with English (Reed-Kellog) sentence diagrams. Facing them to the left this way with the particles on the stem keeps things much more readable.

Don’t add imaginary modifiers

[This overlong section explains why it’s okay to add imaginary subjects, but not okay to add imaginary objects.]

As we described earlier, English is an “SVO” language. Not only does the verb go in the middle instead of at the end, English rules really encourage us to include an object, even when it isn’t strictly necessary.

Imagine this exchange between an English-speaking father and son:

Father: “Who ate the last cookie?”

Son: “I ate it.”

Notice the word “it” at the end of the son’s sentence.

Adding that word would be natural and automatic. Given the context, an English-speaking son would never reply with just the two words “I ate”.¹ He’d automatically add the word “it” without conscious thought. Yet the question of who is fully answered with just “I ate” — the object is irrelevant.

There are occasionally contexts where we’d use just “I ate” in English (after being asked, “Did you eat?”, for example). Most of the time, though, we automatically, unconsciously, add at least a pronoun as the object.

But idiomatic Japanese often omits the object!

In this exact same scenario, a Japanese son might reply with:

「私が食べた」

Note that the object “it” is not in the original sentence, but the English sounds weird without the word “it”.

In Japanese, it would actually sound a little weird (not wrong, but weird) to include the object. It would sound wordy and weirdly explicit for the son to say:

「私があれを食べた」

or

「クーキーを食べた」

when just

「食べた」

suffices. Adding 「あれを」or 「クーキーを」would sound weird even though it is exactly equivalent to “I ate it”. We already know we’re talking about a cookie, so why bother adding an object? Or even a subject? Since the son is talking, the father would know who did the eating!

How should we diagram the son’s reply 「私が食べた」?

English speakers might be tempted to do something like this:

私が食べた

Bad example — DON’T DO THIS

But adding an entirely imaginary modifier and particle when there wasn’t one in the original sentence is incorrect.

A diagram must not invent things that might not have been intended. Sentence diagrams must not change just because the surrounding context changes. The identical diagram must apply whether the sentence is responding to “Who ate the last cookie?” or to “Did you eat?“. No object would be implied in the latter case.

This sentence should be diagrammed as simply “I ate” since that is all that’s in the original sentence.

It’s English rules that make us want to add an explicit object, but we are diagramming Japanese sentences, so “objectless” sentences are perfectly acceptable and common even with transitive verbs.

Both Japanese and English clauses always have subjects, but Japanese sentences especially omit objects quite often.

It’s correct and necessary to add imaginary subjects because every 文節 must have a logical subject that we want to make visible, but don’t add imaginary objects or other modifiers to a Japanese sentence diagram. Only diagram the modifiers that are actually present in the original sentence.

Other modifiers

Modifiers can affect the subject as well as the predicate. They can even modify other modifiers:

白い犬が家の中で静かに食べた

[The] white dog ate quietly inside the house

Notes:

• The core of this longer sentence is just 「犬が食べてた」.

• Modifiers hang below the things they modify in the same order as they appear in the original sentence. The sentence starts with 白い so it attaches to the left of the subject.

  Explicit particles in Japanese allow modifiers to be moved around at will in a sentence, but in a diagram, they always hang from the things they modify.

  The sentence 「家の中で白い犬が静かに食べた」has precisely the same semantic meaning as the original, and would be diagrammed identically, despite swapping the order of things. The repositioned 家の中で modifier would still hang below the 述語 since that’s what it modifies.

• We don’t attach a particle to the stem for the modifier 白い. 形容詞 can also modify 体言 as well as acting as the 述語, but they don’t need a particle. They always confer a sense of nature or state. So the stem of the first modifier remains bare.

  (The word 白い acts as an “adjective” here, but it’s still better to consider it an example of the 形容詞 part of speech. It’s just a modifier like all the others.)

• All of the other modifiers have explicit particles indicating what type of modifying/qualifying is happening.

  The location is “inside” (indicated by で).

  The eating is happening “quietly” (indicated by に).

  Western grammarians tend to think of particles like に here changing the grammatical function of the “な-adjective” 静か from an adjective into an adverb.

  The author finds it easier to think of 静かだ as always and forever 形容動詞. It doesn’t matter whether it used in a predicate or if the だ is swapped out for another particle to indicate some modifying function. It’s just a 形容動詞 specifying a state or condition as always.

  This is the Japanese way to think about the grammar.

• Multiple modifiers affecting the same thing are placed next to each other (中 で and 静かに both affect the predicate, so they are next to each other on the same row).

• Modifiers can also “stack” (modifiers can modify other modifiers). The noun 家 (いえ, meaning house) is modifying “inside”. The particle の indicates which particular inside we are talking about.

• Each row of modifiers adds a little more information to the core clause, but still forms complete, grammatical sentences:

  • The core clause reads 「犬が食べた」

  • With the next row of modifiers, it reads: 「白い犬が中で静かに食べた」

  • With the final row, we get back to the original sentence: 「白い犬が家の中で静かに食べた」

Even verbs can be 修飾語 (modifiers)

Here’s an interesting example:

歩(ある)く習慣(しゅうかん)を身(み)につける

[@] acquires the walking habit

Notes:

• The core clause is simply 「つける」(normally this means “[I] attach”, but here it’s closer to “[I] acquire”). The colloquial expression 「身につける」 means to develop a habit. Literally, it means attaching something to your body (effectively making it mindless and automatic).

• Adding one level of modifiers, we get 「習慣をつける」 or “[I] attach custom/habit”.

• But it’s not just any custom/habit, it’s specifically the habit of walking. Note how the modifier here is a pure 動詞 (verb), 歩く.

Here’s an even more interesting example of using a verb as a modifier,

早(はや)く起(お)きるように努力(どりょく)する

[@] endeavors to early-wake

We’d translate this into English as something like “I endeavor to wake early”, but structurally that is very different than the literal Japanese. This English translation is a compound sentence that uses “wake” as a verb.

In the Japanese sentence, 起きる (wake) is a modifier, what we’d call an “adverb” in English, affecting 努力する (“to endeavor”) which itself is a 動詞 (a verb, albeit a special kind of “suru” verb).

Notes:

• The core clause is simply 「努力する」(”[@] endeavors”).

• 「起きる」is a 動詞, a verb, but with the「〜ように」here, it’s acting as 修飾語 (a modifier, effectively an adverb).

• Further, 「早く」is also acting to modify/qualify the verb 起きる.

Conjugations/inflections

Japanese grammar has an amazing number of ways to conjugate or inflect verb stems.

It’s all quite logical once you get used to it, but you can combine the rules in many ways to fold, spindle, or mutilate the dictionary form verbs in amazingly complex ways.

From a diagramming perspective, though, with few exceptions a verb is just a verb, no matter how complex it becomes.

English doesn’t have a single word that means “did not want to be made to read”, but Japanese does! (To be more precise: Japanese has a way to conjugate/inflect an individual verb to mean that, but English requires several additional words and sentence restructuring to communicate the same idea.)

本を読まさせられたくなかった

[@] did-not-want-to-be-made-to-read [a] book

That ridiculous verb is still just a single inflection of a single verb with the plain form 読む (to read).