Back • Return Home
← Previous • Next →
Lesson 88: The Indestructible Core
[Video Link]
Today we're going to talk about the core fundamental structure of Japanese and a problem that troubles some people about it, which I've tried to lay to rest but there's still some difficulty, I think, so I'm going to try to tackle the question, for what I hope will be the last time, today.
And in doing that we're going to look at a puzzle that's posed by the title of a very popular work.
It's been a manga, an anime, a live-action movie: "Kimi-no suizou-wo tabetai", which means literally "I want to eat your pancreas", and we'll talk about why it literally means that in a minute.
I didn't actually know what a pancreas was.
I thought it was a railway station in London.
But apparently it's a component that you find inside the human body and things don't go so well without it.
I'm not very clear on what you find when you take off the front and back panels of a human body, so we learn something new every day!
That's why this is classified as an educational channel, I suppose.
So, the question that will be raised by this title in the minds of anybody who's got some grasp of the real structure of Japanese, is: Why is the particle -wo, rather than the particle-ga, used here?
If we want to say what in English would be "I want to eat that bread" or "I want to eat that cake", we would say "pan-ga tabetai", "keeki-ga tabetai".
And, as we see, the adjective "tai", the adjective of desire "tai", is pointing not at me but at the cake.
The cake carries the ga-particle, so that is what the adjective is describing.
And confusion ensues when we actually translate this as literally meaning "I want to eat cake", because that's not what it means.
It means "cake is want-inducing (to me)".
But we also know that this adjective of subjectivity, like other adjectives of subjectivity, like "kowai" (frightening), and also potentials like "dekiru" or "taberareru" -- because potential is also a kind of subjectivity.
It's something peculiar to the individual, whether they can or can't do a particular thing.
It's not inherent in the thing itself.
That would be "kanousei".
We know that all of these generally point at the thing that's possible, the thing that induces the desire to eat, the thing that's scary, etc.
But also the polarity can be flipped.
And it's flipped particularly when there isn't an actual cause, an actual visible or tangible cause of the subjectivity.
So if we say, "o-naka-ga suita hayaku tabetai", we're saying "Tummy is empty, I want to eat soon".
Now that "tai" is pointing at me, not at any particular thing, like cake or anything else.
Now, this is the point that a few people have actually resisted and said, "Well, can't we say that it's not really saying 'I want to eat', it's saying 'food in general is making me want to eat'?"
Well, it isn't actually.
It's your empty tummy that's making you want to eat.
And we're going to look at some constructions today, and the one we've just talked about is one of them, that make it a 100% clear
that not only in cases where there isn't a cause of the subjectivity, but in some cases where there is, the adjective of subjectivity can still flip its polarity.
Now, why do people resist this idea?
As a matter of fact, it's not unknown even in English.
We can say "We were happy that day", in which case the adjective "happy" is pointing at us (we're the ones that were happy) or we
can say "That was a happy day", and now the adjective of subjectivity is pointing at the day, which is the cause of our happiness.
We can say "I am suspicious of her behavior" and the adjective "suspicious" is pointing at me (I'm the one who's suspicious) or "her behavior is suspicious", and now the adjective "suspicious" is pointing at her behavior as the cause of my subjectivity.
So this isn't something that doesn't happen, even in English.
For the reason, I think, that people become so agitated about this and so determined to find quite unlikely ways of wriggling out of it is perfectly understandable.
It's because they may have spent months or even years in this terribly confusing world where particles just change their meaning depending on which side of bed they got out of that morning.
So, "-ga usually marks the subject of a sentence but it can also mark the object of a sentence as in 'pan-ga tabetai', in which obviously the bread isn't the subject of the sentence; it's me, 'I want to eat bread'."
Well, we know that this isn't the case.
We know that in those cases it's the bread that's the subject; it's the bread that's making me want to eat.
And if we start saying "That the polarity can flip, doesn't that either break the model of Japanese structure that gets rid of all
these ambiguities or introduce new ambiguities so, ooh there are special rules that sometimes it points this way and sometimes it points that way", well, the answer to that is no, it doesn't matter to the model.
It's quite irrelevant to the model.
Whether we choose to say "pan-ga tabetai" (bread is making me want to eat) or "pan-wo tabetai" (I want to eat bread) doesn't matter.
The only thing that matters to the model is that the particles are always doing the same thing.
If we say "pan-ga tabetai", we're saying the bread is making me want to eat.
If we say "pan-wo tabetai", we're saying that I want to eat bread.
And the model doesn't care which way we say it.
The model is doing the same thing either way.
All the particles are doing exactly the same thing.
Now, can we justify this grammatically?
So if we say, for example, "pan-wo tabetai", surely the problem here is that with "tai" we have an adjective, so we have an adjectival sentence, and an adjective, as we know, can't take a direct object.
So how can we say "pan-wo tabetai"?
And the answer to this is really very simple.
Japanese, as we know, is very adept at gluing verbal elements together to make them into one element or taking them apart at will.
And what's going on in a sentence like "pan-wo tabetai" is that "tai" is no longer being attached simply to the verb "taberu".
We're not saying "pan-wo" and then "tabetai", we're saying "pan-wo tabe..." and "tai" is being attached to that entire unit.
What we want is the action "pan-wo taberu", so we can attach the "tai" to that entire unit.
That's what makes sense of these constructions.
And while "pan-wo tabetai" is a less common way of putting it, and in general "naninani-wo tabetai" is the less common way of putting it, there are some kinds of sentence in which it is the one that we always use.
For example, "tasuketai" (want to help) or "mamoritai" (want to defend).
We don't say "Sakura-ga mamoritai", we say "Sakura-wo mamoritai".
And if we're talking about "seigi" (justice) or "heiwa" (peace) or even "kuni" (country), it's the same.
We don't say Sakura is making me want to defend her, we don't say the country is making me want to defend it, we don't say justice is making me want to defend it, or peace is making me want to defend it. We always say "I want to defend Sakura", "I want to defend justice", "I want to defend peace", "I want to defend the country". Why is that?
Well, essentially I think the reason is that we are not talking about an impulse desire.
If we look at bread and we want to eat it, "ah, pan-ga tabetai" (bread's making me want to eat it).
But we're talking here about things that are more abstract, things that are less impulsive.
And in the case of people, it's more respectful to say "Sakura-wo mamoritai" than "Sakura-ga mamoritai", because we're not saying that Sakura is an object that makes me want to defend her, like a piece of bread that we might want to eat, we're saying that the action of defending her is something I want to do, which is more dignified in the case of a person.
But in the case of, say, the country, or peace, or justice, or even a house or a park, we're talking about something less impulsive and more a decision of our own, if not a conscious decision, then a mindset, a way of thinking that is ours.
So we talk about that action as our own rather than something induced by the external cause.
If we do say "pan-wo tabetai", this is likely to be in conditions where we're talking a little bit more generally, we're not talking about that particular lovely smell of bread by which the bread is making us want to eat it.
We're not talking about some candy we've just seen that's making us want to eat it.
We're talking about the general desire to eat bread.
So we're more likely to say "pan-wo tabetai toki" (when I want to eat bread) than "pan-ga tabetai toki" (when bread makes me want to eat it).
And this isn't a firm and definite rule but this is the kind of tendency, the kind of reason, the nuance, which decides which way we're likely to flip that adjective of desire, in this case, "tai".
So if we come back to "Kimi-no suizou-wo tabetai", why is -wo being used here?
Well, essentially because again this is not an impulse eat.
We're not looking at the girl's pancreas and thinking how delicious it is.
That would be rather hard to do anyway.
Something subtler, something more abstract, something deeper, is being talked about here, the desire for some emotional and quite complex reason to eat somebody's pancreas.
So I hope this time we really laid to rest the final question of polarity switching.
It does not threaten the model.
And it does happen.
And it happens for reasons which are a little subtle, that will probably take time to ingest and immersion to ingest, because you can't learn everything through raw structure.