How do I start to piece it all together?

I never said SOV morphology was rare tho, i said that STRICTLY head final was rare, these are not the same.
In the first place i avoided saying it was SOV because its much less precise, and not very useful knowledge compared to saying it is head final. Because japanese uses particles to indicate grammar, theres no need to talk about a syntax categorization because the only syntax that matters is that the head of the sentence comes in last.

telling a beginner that japanese is SOV is not nearly as helpful as telling them that the head of a sentence is always last, the former only tells us about the case where a sentence has a verb, but what if it is an A-is-B type sentence without a verb, then what ? then knowing that the head is always last is much more helpful to us. Not to mention, theres no rule that the subject must come first in japanese.

In contrast, english is a mix of head initial and head final depending on the sentence.
Head final is a golden rule for japanese which can really act as a shining beacon for japanese learners because it is consistent and always true.

Dunno about this. Head-final is definitely more true, but then you have to try and explain what “head” and “head-final” actually mean.

Side note: This reminded me of word order in Turkish. I’m not sure if Japanese is quite as flexible though.

2 Likes

そうだね~…

…is not head-final. I agree that head-final phrase forms are fundamental to standard Japanese syntax, and once that clicks everything gets easier. Sentence-tagging particles like ね and ぞ and な and っけ are a crucially important exception to Japanese’s head-final grammar and worth pointing out.

In case this whole discussion is going whoosh over your head, have a nice big post :slight_smile:

Japanese use a word order where a key significant word called a “head” comes after everything else, where in English it tends to come towards the beginning except when it doesn’t. Sometimes English lines up with Japanese pretty well.

Japanese makes a distinction between attributive (“stuck-to-a-noun”) and predicate (“not-stuck-to-a-noun”) verbs and verb-like adjectives and noun-like adjectives and nouns which are moonlighting as adjectives. Since this is a good way to show what head-finality means and cover how different parts of speech stick to other parts of speech, and introduce some conjugation too, and this whole idea of embedded phrases and recursion… let’s jump in and piece some things together for fun.

The heads for the examples are in bold. See if you can get a feel for how everything leads into the head…

So here’s a verb-like adjective everyone knows: かわいい (cute) kawaii.

  • mouse is cute - ネズミがかわいい。nezumi ga kawaii

かわいい isn’t stuck to a noun and it’s at the end of the sentence: it’s a predicate and everything else leads up to it. かわいい is the head, everything leads up to the head.

かわいい is also in plain form here. Does it need です desu to mean “it’s cute” because です means “is”? No. This is fine. If you add です, what you’re saying is…

  • mouse is cute (and I’m saying so politely) - ネズミがかわいいです。nezumi ga kawaii desu

です here doesn’t quite equal “is”. It means “mouse is cute (and I’m saying so politely)”.

Japanese gets taught to foreigners using 丁寧語 teineigo - a polite form of Japanese which strikes a balance between pleasant and uncomplicated. Teineigo means sentences tend to end in です desu or ます -masu, depending on what’s at the end of the sentence.

A plain verb-like adjective at the end of a sentence is made teineigo by popping です desu after it, and verbs are changed so they have ます at the end. (Let’s not talk about past tense or negation or verb groups just yet.)

How about when we want to say the cute mouse is doing something. First we need to talk about how to say “cute mouse” - how to give the mouse an attribute:

  • cute mouse… - かわいいネズミが… kawaii nezumi

We just put the plain adjective at the front - this isn’t a full sentence, just a “noun phrase” which is to talk about a thing. The key word of noun phrases - the “head” - is a noun, so with Japanese being “head-final” that means the noun comes last in its own chunk of the sentence.

Since we’re talking about a mouse who is already cute and say that it’s running, so we have to give ネズミ the attribute of かわいい - since it’s attributive, we use the plain form of かわいい.

Now we can say what the cute mouse is doing…

  • cute mouse is running - かわいいネズミが走っている。kawaii nezumi ga hashitte-iru

To say that a verb is ongoing (is running as opposed to runs), we put the verb in て form and add いる afterwards. Since いる is last, it becomes the “head”-like bit.

What’s て form? It is one of these big topics in Japanese grammar that is too big to explain in a paragraph. For now, it’s enough to know that verbs and verb-like adjectives both have a て form, and words in て form rely on something that comes afterwards to paint the complete picture - including whether the sentence is in the present or the past.

So to act as predicates and attributes, verbs in て form need some extra glue. Here, いる is the extra glue and put together it means “some action is ongoing” or sometimes “something has come to pass”.

Now because いる is acting as the “head”, we make changes to いる - if we want to be polite we turn いる iru into います imasu to get 走っています, If we want to put this running in the past, we say 走っていた hashitte-ita.

Now how about the other way around - what if we want to single out a mouse which is already running and say that it’s cute?

  • running mouse is cute - 走っているネズミがかわいい。hashitte-iru nezumi ga kawaii

We swap them over. That’s how you stick verbs and verb-like adjectives to a noun - put them plain form then just stick them straight on. Now hopefully you can see why these verb-like adjectives are verb-like - they behave a lot like verbs.

And just like before…

  • running mouse… - 走っているネズミ… hashitte-iru nezumi

…everything leads up to the head.

Now, what if this mouse is hungry…

  • cute mouse is eating bread - かわいいネズミがパンを食べている。kawaii nezumi ga pan wo tabete-iru

“Bread” doesn’t come at the end in Japanese because verbs are the “head” of sentences. Everything leads up to the head in that bit of the sentence. In polite form, we change the いる to います because it’s at the end of the whole sentence in the predicate position. We don’t change かわいい because it’s an attributive and has to stay in plain form.

So, what if the mouse isn’t just cute but funny - 面白い

  • the interesting and cute mouse is eating bread - 面白くてかわいいネズミがパンを食べている。omoshirokute kawaii nezumi ga pan wo tabete-iru

Here we put 面白い into it’s て-form (continuative form) to grammatically indicate “this but also another thing”.

  • cute mouse is eating bread - かわいいネズミがパンを食べています。kawaii nezumi ga pan wo tabete-imasu

Now, what if we want to say…

  • the mouse that is eating bread is cute - パンを食べているネズミがかわいい。pan wo tabete-iru nezumi ga kawaii

If it’s not immediately clear: you use the “head final” structure to back up entire attributive sentences onto a noun in Japanese, with the verb in plain form.

Can we て? Maaaaaybe?

  • the mouse that is running and eating bread is cute - 走っていってパンを食べているネズミがかわいい。hashitte-itte pan wo tabete-iru nezumi ga kawaii

Notice how we stack up the て for “hashitte-itte” since いる iru is carrying that final “head” position. This is stretching the boundaries of natural sentences a bit but sure, て all the things.

So what about nouns and noun-like adjectives? Let’s start with predicates…

  • it’s a mouse- ねずみ。nezumi da

Or in polite form…

  • it’s a mouse (and I’m being polite) - ネズミです。nezumi desu

です is the polite form of だ equivalent to verbs which end in ます. Here it’s much more like “it’s”.

Where does this です even come from? It’s a contraction of であります - the plain form of あります is ある, so the plain form of です is である. である is used a lot where you might expect だ or です in formal/scholarly/officious situations like dictionary definitions or legal documents. Hopefully it is easy to understand how である and であります are the same idea at different levels of politeness.

And the で before である and ではありません is just the て-form of だ.

  • it’s a mouse (and I’m writing a dictionary definition or contract in which this is some kind of stipulation) - ネズミである。nezumi de-aru

Noun-like adjectives follow the same pattern as nouns…

  • it’s quiet - 静かだ。shizuka da

And again, for politeness…

  • it’s quiet (and I’m being polite) 静かです。shizuka desu

The thing to note here is that both nouns and noun-like verbs both need the help of だ (or です) to be predicates, because unlike verbs and verb-like adjectives they can’t be the heads of a sentence without that stub.

Onto attributives, and here we’re just talking about a noun phrase where the noun is the head…

  • a quiet mouse… - 静かなネズミ… shizukana nezumi

To be the attribute of a noun, a noun-like adjective needs help from な. Which sounds an awful lot like だ, doesn’t it? (Some people liken な to a sticky attributive version of だ.)

Nouns can stick to things too. You may already know about の.

  • a cat’s mouse… - ネコのネズミ… neko no nezumi
  • it is the cat’s mouse - ネコのネズミ。neko no nezumi da
  • it is the cat’s mouse (and I’m being polite) - ネコのネズミです。neko no nezumi desu

Particles can also form their own phrases as part of a sentence, and again in Japanese we know that words lead up to the head. Some particles are what Japanese uses to express ideas like “from”, “to”, “until”, “especially”…

  • from/because of the mouse - ネズミから… nezumi kara

And because a particle phrase needs to live in a sentence and phrases in Japanese lead up to their head, we get this.

  • i/we/you left because of the mouse - ネズミから出た。nezumi kara deta
  • i/we/you left because of the mouse (and I’m being polite) - ネズミから出ました。nezumi kara demashita

Oh, another sentence… say, can we do that thing from before even if it’s in the past tense?

  • it’s the person who left because of the mouse - ネズミから出た人。nezumi kara deta hito da

If we use the plain past verb we can absolutely do that thing from before. But what if it’s past tense?

  • it was the cute person who left because of the mouse - ネズミから出た人だった。nezumi kara deta hito datta

だった is the past form of だ if that’s not clear. But what if the person was cute? We have to put cute in the past tense too with -かった -katta

  • it was the cute person who left because of the mouse - ネズミから出てかわいかった人だった。nezumi kara dete kawaikatta hito datta

Or what if everything was cute.

  • i wonder if it was the cute person who ran around in a cute way because of the cute mouse… - かわいかったネズミからかわいく走り回っていた人だったかな。kawaikatta nezumi kara kawaiku hashrimawatte-ita hito datta kana…

Wait, かな isn’t the head? No. It’s a “sentence-tag” particle, not a head, and that’s the big exception to an otherwise steadfast pattern of everything leads up to the head.

Hopefully you can see from all these examples the sheer consistency of head-final phrases in Japanese and have a sense for how embedding these phrases within one another works - . Beyond that, being proficient with Japanese grammar is being able to see how one class of word sticks to another class of word - especially where particles are concerned - and knowing the difference between “sticky” attributive forms and “end of sentence” predicate forms.

I have left a lot out of this explanation and glossed over a ton of detail in order to keep the focus firmly on head-finality, but hopefully the head-finality bit is now as demystified as it can be.

Good luck!

12 Likes

I dont think that is actually a proper grammatical phrase but simply colloquialism doing its thing.
I usually argue that sentence ending particles (and other non-grammar particles) are not grammatical constructs and as such do not take anything away from the head finality of the language.
Of course a beginner should be informed of this and ideally that would happen with an introduction to what particles are and the types of particles (with 1 being emphasis/sentence ending particles)

Aye i knowingly skipped that step in order to keep it short(I want to avoid writing a lesson), thankfully its fairly easy to look up.
I may write my idea of a short introduction paper on japanese at some point for the heck of it, and i’d definitely include an explanation of it there.

Agreed, certainly a typical beginner would have no idea about those, unless they’ve studied linguistics. I had to spend some time understanding what all that jargon actually means, although a lot of the things I had gotten a feel for intuitively.

Interesting tidbit about Turkish. Finnish is actually very similar.

Regarding “headness”, Finnish seems to be a mixed language where you can pretty much use both. For example:

1* Kirja, jonka lainasin kirjastosta. = A book that I borrowed from the library
2* Kirjastosta lainaamani kirja = [from the library borrowed] book

Number 2 resembles a lot the Japanese head-final structure, where the clause in front modifies the noun as a whole. I can’t really think of a way to say the same thing in English. I still had to wrap my head around thinking this way constantly, so I can imagine how difficult it is for native English speakers, who don’t have this kind of structure at all!

but abstract enough to be much harder to understand than what a subject or an object or a verb is - I can give a simple one sentence explanation of each of those that allow you to pick them out in a sentence.

I understand what a head is as a concept but I’m not confident I could reliably pick it out in practice. And I certainly couldn’t give a simple explanation on how to.

@morteasd That’s cool, I didn’t realise Finnish was agglutinative… Turkish is considered strictly head-final though, and I’m not sure how to exactly relate that back to variable word order…like I said, “head” is a comparatively difficult concept :sweat:

1 Like

そうですか。

That’s definitely not colloquial. That’s polite Japanese.

か is actually pretty fascinating in and of itself when you find it in stuff like 何か and 誰か, but it’s definitely an end of sentence particle and definitely not the head of a phrase.

On a more serious note, real actual linguistics doesn’t let you handwave something as “colloqualism” (whatever you intend for that to mean) on the basis it doesn’t fit a conveniently simple theory. It’s not very… rigorous.

I’ll correct myself, that is indeed a valid phrase but let me explain why its definitely still head-final

そうですか works because そう refers to something previously mentioned which if you were to spell it out would present you with a gramatically complete sentence, in which you would be able to see the head-finallity clearly. Japanese is known for liberal omissions but that doesnt change the structure of it.
you would use そうですか in a situation where someone else has just given you some information and you respond with “is that so?”, rather than repeating everything just said and making it a question, so instead it is omitted.

Whhoooaa! I came here thinking I’d pickup some advice I might have missed before but talk about getting more that you bargain for, heh?

Full blown WK community launching into liguistics!

Hope you’re enjoying the read @Try4ce

I purrsonally stayed for the miiice! :cat2: :cat2: :mouse2: :rat: :cat2: :cat2:

5 Likes

Quoting Wikipedia:

In linguistics, the head or nucleus of a phrase is the word that determines the syntactic category of that phrase.

By my naive understanding of that definition, the か should be the head, as it is what determines the category of the phrase as a question (informally - I don’t know what it would be if you properly analysed it).

After reading around a bit, I get the feeling that there are a couple of competing definitions of what the “head” means, but that the one I quoted is the appropriate one when talking about the head-directionality parameter (what a pompous name…).

Definitely this.

I’d disagree. The former is merely a “Lies To Children” handwave that’s digestible enough to get started.

It’s the “lies to children” thing again where the copula is explained as a verb that means “to be”.

Now, I do think that you need to get into those details sooner rather than later, but you have to start somewhere.

1 Like

そうですか really isnt a very good example for explaining what a head is because you cannot actually see the head. (It also doesnt really mean anything without a context)
It’s like the case of a subjectless sentence. It’s not because theres no subject, its simply that the subject has been omitted because it was already clear from context.

You need to define head and head-first/last properly, because I don’t think you mean what I understand by them.

For Japanese it’s simply the last Noun, Verb or adjective of a sentence.
The resultant sentence can either be an A-is-B or A-does-B.
It is essentailly the main state, or main action of a sentence.

For OP: This post is probably skippable. ごめんなさい。orz

Ah. Uh. No. Linguist to the rescue again. Question is not a category of phrase for the purposes of head-marking.

Phrases here are noun phrases, verb phrases (often equivalent to sentences), adjective phrases, adverb phrases, etc. They centre on parts of speech which they are named for.

So think of it as “a chain of words which can syntactically sit in a sentence as a chain of words for that kind of phrase”. If you can substitute another phrase in and the sentence still makes sense in the same kind of way, it’s the same kind of phrase.

So

  • a dog/いぬ is a noun phrase
  • that dog/その犬 is also a noun phrase
  • that red dog/その赤い犬 is also a noun phrase
  • that big friendly red dog who barks at the mailman all the time is also a noun phrase

How to check? Again, sub it in: “a dog is drinking water” and “that big friendly red dog who barks at the mailman all the time is drinking water” are both syntactically valid, even though structurally they’re not. You can embed other phrases within noun phrases because grammar is componential - it lets you do that. So “that big friendly red dog who barks at the mailman | all the time | is also a noun phrase” contains a verb phrase for attribution (barks at the mailman all the time) which in turn contains a prepositional phrase (at the mailman) and an adverb phrase (all the time).

As I was taught, “verb phrase” is often synonymous with “sentence” since sentences have a verb at their core. If a sentence in Japanese is a verb phrase, and Japanese phrases are quite regularly head-final, then at least while it’s the predicate of a sentence there’s some flexibility as to whether the head is literally the final word of the verb phrase.

But I can check a book or two to see if they agree with me since I’m almost certain Japanese linguists have their own notions and norms about this sort of thing.

For OP: You can skip this one too. Sorry I’m doing these from the bottom of the thread up! :slight_smile:

The head is the copula です.

1 Like

Could I just propose that ‘heads’ are usually what defines the ‘part of speech’ of a phrase? I think that’s the sort of ‘syntactic category’ mentioned in the Wikipedia article, based on the examples given. The problem with か that it usually acts as a question mark or a tone-softener when added at the end of a sentence, so I don’t think it can be called a head in sentence-final position. (The other problem with か… is that I can’t find any phrase ‘head’ analysis on a sentence that includes it…)

I think, however, that か can act like a head in sentences like 今行くかわからない. In that case, 今行くか becomes a sort of relative clause that is syntactically the object of わからない, and that’s thanks to か. Without it, speakers would be obliged to nominalise the phrase in some way so that it could act as the object of わからない.

While it can certainly be viewed that way and may be true theoretically (to the extent linguists agree about it). I don’t think its relevant to look for the head in such a phrase, as its entirely dependent on context.
And i would still argue that it’s a grammatically incomplete sentence since it has no meaning by itself, but the lines are getting blurred at this point.

Regardless, it takes nothing away from the concept of head-finality.

I was pretty sure it wasn’t :stuck_out_tongue:

But so, what even is か? I get that “question” can’t be a syntactic category in English, but I’m not convinced that extends to languages that do have explicit syntax for questions (e.g. Japanese, Chinese, Turkish). As @Jonapedia says, I haven’t found any examples of this kind of analysis for questions.

For OP: The particle か ka is very mysterious. Keep an eye out for it. :slight_smile:

True, but that’s not a predicate use - more of a a nominalisery particley use for marking when something is unknown or unresolved. Sentence-medial か is still a bit of an enigma for me, intuition-wise.

Another completely unconfirmed intuition I have is that the か in 誰かです。 and 誰ですか。are the same か because all that uncertainly certainly feels the same…