I’m sure we’ve all struggled with が in the past (I know I have and continue to do so). It’s one of those things you’re told you’ll understand after enough exposure, and while I’ve come to understand it’s role in a sentence as a subject marker, the thing that bugs me is the role of “introducing” new information.
So here is my question:
What classified as new information, and what classified as universal information?
This has been my one big hang up with が, and I was hoping someone could help me out here.
Here’s 2 pages from Chapter 2 of The Structure of the Japanese Language by Susumo Kuno. I thought it would be answered succinctly, but he keeps expanding so no matter where I stopped taking pictures there would be a cliffhanger for the next page, all the way until the end of the chapter.
It goes into more linguistic detail than you could ever want about plenty of topics (check the look inside for the table of contents). And it has 2 chapters alone on the difference between は and が.
Universal Information is something you and the people you’re talking to already know.
Let’s say you want to talk about Japan. If everyone already knows Japan, it’s universal and you can add new information like “I went to Japan last year”
If they don’t know Japan, you can say “There is a country called Japan”. In that case, “country” would be universal information and Japan new information.
I wholeheartedly agree with the recommendation to read Kuno’s chapters on the question. For those interested in a summary thereof, Martin gives one in his Reference Grammar, section 2.3 (typography as in the original):
What this helpful description of Kuno’s boils down to is that N ga in all sentences MAY localize the emphasis (so that the question is about N, not S) and in certain situations can ONLY do that, as in the unsubordinated identificational sentence; but in many situations it is not required to do so: most sentences with N ga are ambiguous as to whether they are answering questions localized on N ga or questions about the sentence as a whole (or localized on some other adjunct), just as sentences with N o are usually ambiguous in the same way–unless the localized emphasis has been preempted by a specifically interrogative adjunct elsewhere or by an N ga that can only be interpreted as ‘N and N only’.
Elsewhere on the same page, Martin lists those “situations” where が only acts exclusively. I read them as follows (any mistake or omission is most probably mine):
Either…
predicate expresses a permanent state and subject is not quantified;
or subject is first person or described as moving or located with respect to oneself.
And…
a preceding が in the same sentence does not already take the exclusive meaning;
Say I’m having a conversation with someone. I can only make a subject a topic with は if either, it has been brought up before in the conversation already, or it belongs to a universal concept that everyone would know about/something based on the context of the situation(i.e. the sun, myself, etc.). If I wanted to bring something up to be able make it into a topic, what counts as a universal topic, and what counts as a new subject?
People’s answers so far have been helpful, but I feel like I’m still missing something. All the more likely is that I’m probably over thinking all this.
Maybe think about how you bring up new topics in English. Let’s say I’m saying “the dog smells bad”. If I’m talking about my new dog that I have at home, my friend is going to be confused and say “what dog?”, meaning I needed to introduce that topic previously. If there’s a stray dog literally snuffling around our feet, then it’s clear to my friend which dog I mean, so I don’t need to bring it up.
So in your example, the dog your friend didn’t know about would be followed by が since you were bringing it up the first time. And the stray dog would be followed by は in this case because you both are aware of it. Right?
If that’s the case, then this is really helpful. For whatever reason I never thought to compare it to English conversations. I guess the best way for me to look at is if I could say whatever I’m talking about as the subject of the sentence and be vague (like The weather is bad), but anyone listening still understood what I was talking about, then it needs to be followed by は. If I said it and they would be confused then it needs to be followed by が.
Well, that’s at least what I’m guessing. Does that sound about right?
I have a theory that は holds similar connotations as “the,” and が hold similar connotations as “a.”
They’re by no means 1 to 1 relationships but maybe comparing them can help make it more intuitive.
For example:
“The story” implies that the listener should know which story the speaker is talking about.
“A story” implies that the listener doesn’t know the story, or if they have, they aren’t expected to be able to guess which story the speaker is talking about without more information.
Watchu think?
You wouldn’t? 天気はいいですね sounds like valid small talk to me. But that’s probably because the person you’re talking to is expected to already know/agree that the weather is good. On the other hand, I think that if they’ve been inside all day and don’t know that it’s raining, が would be more appropriate - like saying, “There’s a bad storm outside,” instead of, “The storm outside is bad.”
Was about to point out the same comparison, but you beat me to it. I think Jay Rubin explains it this way as well in his Making Sense of Japanese.
At a glance, I think saying 「天気はいいです」 would come across as introducing the weather being good as a new topic, by contrast to whatever the previous topic was (i.e., implying that the previous topic was not so good)? Whereas using 「天気が…」 would be more what you’d say if you were making a random small talk observation about the weather?
Not related to the topic at all, but I happen to be reading this book as well, and it’s very good. I was curious if either of you read the chapter on stative/non-stative verbs? It’s about how to interpret non-past tense and ている forms.
I don’t know how to describe it but his conclusions in that chapter are pretty crazy, if they’re true. However, I’m not sure whether I should put stock in them because I can’t find any sources in English that corroborate it or refute it. Just wondering if either of you have any insight on that.
天気はいい sounds like the speaker is contrasting something. Perhaps they or the person they are talking to are having a bad day, but in contrast to that, they comment on the fact that it’s sunny and nice.
I think that the は vs が issue stems from how many resources imply that it can be one or the other. In other words, at certain times both mark the subject; but that’s not correct. The confusion lies with the properties of Japanese allowing the subjects to be omitted. In the absence of the subject, learners are under the impression that は does the job for が instead. Mainly because textbooks usually introduce は first and が later. Because the subject can be assumed by the topic or the context of the situation, its presence in such cases are deemed redundant.
Having a strong grasp about who or what is the subject, it makes parsing out why universal information is marked with は and new information is marked with が. Really the reason why distinction is made could be because people aren’t sure how to use those particles properly to begin with.
For example,
ゾウは鼻が長い
Elephants’ noses are long
This sentence is not presenting new information to anyone who has seen an elephant before. Yet が is used because nose is the part that the speaker pointing out to being long. は is used to focus the listener on what the speaker is going to say, essentially a property that may be unique to elephants compared to other animals.
But if something is out of the ordinary or ambiguous without clarification, が can be used to point out who or what it is.
So in the case of bring up something new, just like in English, you have say something to broach that topic if it’s not the focus of the conversation.
One example, a person couldn’t come to this week’s rehearsal for a performance because they are sick might not even use は at all.
今週の稽古なんですけど。体調が悪くて休ませていただきます。申し訳ございません。
About this week’s rehearsal, I’m not feeling well so I’m going to rest. Sorry everyone.
or using は
今週の稽古は体調が悪くて休ませていただきます。申し訳ございません。
About this week’s rehearsal, I’m not feeling well so I’m going to rest. Sorry everyone.
Of course, these are things I’ve learned from exposure over time, but figuring out what is the subject helped clarify the new information vs universal information thing for me because は and が’s functions are distinct.
I’ve read the chapter, but I’m not sure what it is that you find so crazy; I’m afraid you’ll have to be more specific.
For instance, regarding non-past tense being interpreted as future only, I don’t think there’s much disagreement about that being a feature of some verbs. Different authors have different classifications and different criteria (e.g., Martin’s Reference, section 3.12, quotes Yamada’s classification (among others), which has an activity class for those verbs and different criteria).
I mean, it’s just what linguists do: find criteria to classify the observed language. It’s highly unlikely that Kuno’s conclusions would be downright untrue, in any case. At worst, his classification might be disputed, might misclassify some things (exceptions), or fail to account for some patterns, but at the core, if he’s done his job properly (which I believe we have no reason to doubt), the basics should be fine.
Crazy was an overstatement obviously. Kuno’s system was nice but regardless of whether he was right or wrong I wasn’t comfortable with the fact that I couldn’t find anything to cross reference it with.
I see. I think Martin is still the classic reference today (for a “traditional” grammar), so it’s always worth cross-checking there. I use it as my starting point for anything Japanese-grammar-related (except for historical queries, for which Frellesvig’s History provides a better entry point IMHO).
The book is hard to find, not easy to search if you’re not used to it, and the formatting is hard to read, though. (That’s a lot of issues right there. :P) But it usually has something to say on most topics, and the system presented is very comprehensive. Other newer specialised sources may provide a more adequate treatment of some topics, though, naturally.
