2

I just read an article explaining that が should not be thought of as a subject particle, since it can also mark the object, which is fine and dandy on its own.

However, I also remember reading an article that explained that when certain transitive verbs (e.g. 話す) get conjugated into particular forms (e.g. potential), anything that would have been marked with を is instead marked with が.

At first, these two bits of information seemed contradictory (the fact that the latter was explained in terms of the object becoming the subject didn't help). In trying to reconcile them, however, I came to the conclusion that を can be replaced with が (with a change in nuance), but が can't (necessarily) be replaced with を. Does this seem accurate, or is one of the sources of information wrong?

themathandlanguagetutor
  • 209
  • 1
  • 6
  • I think it's more accurate to say that Japanese happens to treat some particular sentence constructions in a different way to how English does, so where we would identify a particular actor as the object in English, in the Japanese sentence it's the subject (and what we would consider the subject is actually something else, potentially the topic). But that's more how I understand things rather than any official statement on Japanese grammar so I'm reluctant to put it as an answer. – ConMan Nov 21 '17 at 05:53
  • Well, one of the points the article I linked was making was that, even when viewed from a Japanese grammatical point of view, が can create ambiguities about what is meant as the subject and object. – themathandlanguagetutor Nov 21 '17 at 05:57
  • Possibly of use: http://www.imabi.net/theparticlegaii.htm – G-Cam Nov 21 '17 at 13:30
  • 1
    From imabi's link, 29. 「俺、クモが怖くないよ」should be クモは. – user4092 Nov 21 '17 at 14:18
  • @G-Cam this article seems to be describing an event where the object in English becomes the subject in Japanese, rather than instances where が accrual marks the object – themathandlanguagetutor Nov 21 '17 at 23:53

2 Answers2

3

From the article:

クレープが食べたいから。

The subject of this sentence is crepe. 食べたい does not mean "wants to eat" any more than 好き means "likes". 食べたい functions as an adjective, expressing that crepes are "eatable" (not in an ability way, but in a preference way). It's no coincidence that 〜たい constructions conjugate exactly like い-adjectives.

So, literally: "Because crepes are eatable."

It's conspicuous that Mr. Kim didn't provide any other examples of が marking an object. To play devil's advocate though, I'll come up with one on his behalf:

私は日本語が分かります。

The most natural way to translate this into English is "I understand Japanese."

"I" is the subject, "Japanese" is the object. Simple.

Unfortunately (or fortunately, depending on your point of view) English is not Japanese.

What's going on in the Japanese is more like this: "As for me, Japanese is understandable."

This is because 分かる is an intransitive verb. It doesn't take an object like "understand" does in English. In Japanese, something "is understandable," or "is clear." All you have to do to verify this is look up 分かる in the dictionary of your choice.

So I would take what Tae Kim says with a grain of salt. Calling が the subject marker is totally kosher.

(Calling は the subject marker, on the other hand......)

Nahcirn
  • 315
  • 1
  • 7
  • I also want to point out that in the comments under the article, Tae Kim actually admits that he got confused: "Your comment was very insightful. たい might be a special case semantically in what can be the “subject”. It might be more accurate to say that が indicates the subject in addition to identifying the unknown." That's a striking contrast to the title of his article, which claims that subjects don't even EXIST in Japanese! No clue why he didn't remove the article at that point. – Nahcirn Nov 21 '17 at 10:41
  • How would you reconcile the claim that if you really *did* want to say that crepes want to eat, you'd still use "クレープが食べたい"? – themathandlanguagetutor Nov 21 '17 at 16:44
  • @themathandlanguagetutor What's to reconcile about it? Ambiguities exist all over the place in Japanese if you don't have context. It has nothing to do with the が particle specifically. 食べた is a complete sentence, but it doesn't tell you a whole lot without context. – Nahcirn Nov 22 '17 at 00:19
  • You're saying that "食べたい" is used as an adjective, and that が really does mark the subject. Assuming these are correct, "クレープが食べたい" could *only* mean "[someone] wants to eat crêpes" and there would be no ambiguity. – themathandlanguagetutor Nov 22 '17 at 00:36
  • Actually I said the opposite. It *can't* mean "[someone] wants to eat crepes." If が marks the subject, then that is "crepes." In the sentence you gave, the subject is "someone" and "crepes" is the object. The reason there is ambiguity is that we don't know what the topic is (but we would, if we had context). – Nahcirn Nov 23 '17 at 05:17
  • The subject would be whatever the adjective describes, i.e. the thing that someone wants to eat. I think you're trying to be too anglocentric. – themathandlanguagetutor Nov 23 '17 at 07:47
  • In addition, even if I have what it can and can't mean backwards, there's still no ambiguity, because one of the meanings is eliminated. – themathandlanguagetutor Nov 23 '17 at 08:06
  • @themathandlanguagetutor Correct, the subject would be crepes, the thing that someone wants to eat. That's what I said. And as I said, the ambiguity comes from lack of context, not some dual function of が. – Nahcirn Nov 23 '17 at 08:21
  • Okay. You seem to be saying two completely different things here. A: that the crêpes are definitely the thing someone wants to eat and not the thing that wants to do the eating. B: that the sentence is ambiguous and we can't tell which it is. – themathandlanguagetutor Nov 23 '17 at 08:30
  • In the former case, the claim I originally said needs reconciling does so because it is obviously false, since が can only mark the subject (which in this case is the thing that wants to be eaten); in the latter, either your entire point is wrong or you explained yourself poorly, because if が is really the subject marker, there would either be no ambiguity, or the ambiguity is because there's an ambiguity in the usage of 食べたい. – themathandlanguagetutor Nov 23 '17 at 08:31
  • @themathandlanguagetutor You're right I explained myself poorly, because your last assumption is correct. 食べたい is ambiguous without proper context. Let's look at 私が食べたいです。This is a contracted form of 私が（何かを）食べたいです。I should have mentioned that 食べたい doesn't *have* to refer to a subject. You can say 晩御飯を食べたい？ to mean "Do you want to eat dinner?" But just because 食べたい doesn't have to refer to a subject doesn't mean that が never marks the subject. I guess you could argue that 晩御飯を食べたい is short for 晩御飯を食べるのが食べたいです。 – Nahcirn Nov 23 '17 at 19:55
  • ...But the point is that it's better to think of が as consistent in marking the subject, and the unstated context as ambiguous. Otherwise you're introducing confusion into places it doesn't exist. – Nahcirn Nov 23 '17 at 19:56
  • That makes sense. Thank you for putting up with me through all of this haha. – themathandlanguagetutor Nov 23 '17 at 21:55
  • 1
    It's all good. For a language as different as English, even the most basic stuff can be puzzling. Anyway, as a general rule: when you have to make assumptions, I find it good to always assume the thing that keeps the language the most consistent. In this case it's that が marks the subject. – Nahcirn Nov 24 '17 at 07:51
0

I think が functions as both a subject marker and a kind of object one, not solely subject.

What Tae Kim calls identifier actually can be explained as emphatic construction. For example, when you rephrase 食べたのは魚だ (it's fish that I ate), you say 魚を食べた (often with stress on を), not 魚が. So, it's not so much function of が itself as the case particle not being topicalized because of emphatic construction.

Next, the reason why could-be objects for potential verbs or the likes are not a subject can be understood from comparison between these examples.

  • 俺が怖い？
  • ワタシ、キレイ？

Given that both of them are question with Sentence of neutral description, thus, the subject is indicated with Zero particle in the latter sentence (the famous question by Kuchisake-onna). However, the first one still has が because it means "Do you find me scary". If it meant "Am I scary?", it would be "俺、怖い？". This shows how aforementioned が is different from the general subject marker one, apart from if you regard it as an object or another subject.

user4092
  • 16,303
  • 19
  • 23
  • I'm sorry. but I'm not sure what you mean by "the case particle not being topicalized"; could you please re-explain that bit. Also, I'm pretty much completely lost after that; I think that might be a bit advanced for me, unfortunately (but I'm perfectly fine just accepting that potential forms are intransitive). And is "俺、怖い？" really the most idiomatic way to ask "Am I scary?" I was under the impression that dropping particles was merely a product of being more causal as opposed to a mechanic to convey meaning. – themathandlanguagetutor Nov 23 '17 at 08:00
  • Zero particle is not necessarily a mere result of dropping it but can be grammatical requirement (1) for the subject in question with SoND or (2) when the subject in SoND is modified with words like この or その. – user4092 Nov 24 '17 at 18:36
  • Unlike normal question, that of SoND is asking if it happens to be the case in the moment. In other words, 私、きれい？ is asking if the speaker looks pretty in the moment, while questions with topic parts like 私って、きれい？ ask if s/he has aesthetic attribution. – user4092 Nov 24 '17 at 18:49
  • "The case particle not being topicalized" means using が、を、に etc instead of は、は、には etc. – user4092 Nov 24 '17 at 18:55