Japanese Language Stack Exchange is a question and answer site for students, teachers, and linguists wanting to discuss the finer points of the Japanese language. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top


Why を and not が twice?

share|improve this question
Just curious: where did you find this sentence? – fefe Jan 6 '12 at 16:10
Related: japanese.stackexchange.com/a/3474/542 – Flaw Jan 9 '12 at 10:54
up vote 13 down vote accepted

This is an interesting phenomenon. In a simpler sentence:


the predicate 好き is a na-adjective, so it does not have the ability to assign accusative case to its arguments. Therefore,


is ungrammatical. However, when it is embedded under a verb that can assign accusative case and there is no other noun phrase that needs to be assigned accusative case, then the object of 好き can receive accusative case (remotely) from that verb. Either the nominative () or the accusative () becomes available.


In the example above, receives accusative case not from its predicate 好き but from the verb 思う. This is called Exceptional Case Marking construction.

Similarly in English, the subject in a simple sentence needs to be in nominative case:

He is smart.

so having accusative case on the subject is ungrammatical:

×Him (is/to be) smart

However, when it is embedded under a verb that can assign accusative case and there is no other noun phrase that needs accusative case, then the embedded subject can receive the accusative case (remotely) from the verb in the main clause:

I consider [him (to be) smart].

Here, him is accusative case marked not by its predicate smart but by the verb consider.

share|improve this answer
Perhaps if your answer's English examples used the same words as the Japanese sentence ("teacher", "cat" and "like") it would be easier to make the association between the languages. (But then again "like" is a verb in English, while 好き isn't a verb in Japanese.) – Flaw Jan 9 '12 at 10:58
@Flaw You are right. It is usually better to try to show the parallel with the corresponding sentences from the different languages. But in this case, the parallel is not a strict one. Whereas in Japanese, the nominative/accusative alternation in question happens with the embedded object, the alternation in the English counterpart happens with the embedded subject. So I couldn't make a strict parallel anyway. And also due to the category difference of the predicate, as you correctly pointed out. – user458 Jan 9 '12 at 11:30
@Flaw Furthermore, I needed to use the pronoun he, him because English nouns do not show case differences. – user458 Jan 9 '12 at 11:37
Thanks, this is a superb explanation! – SLC Jan 9 '12 at 16:45
@sawa I think it can happen with the embedded subject in Japanese also: 「私は先生を天才だと思う」. With a 形容動詞 predicate you can say that 先生 is the (nominally-marked) object, but in a simple copular sentence like that, 先生 is certainly the subject. Following this train of thought, perhaps it's not even a question of subjectality/objectality but rather just whatever was nominatively-marked before the embedding -- I think some speakers accept sentences like 「私は先生を腕を長いと思う」 for that reason. – Darius Jahandarie May 22 '13 at 18:52

私は先生が猫を好きだと思います sounds all right to me, while 私は先生は猫が好きだと思います would sound more natural.

Actually the first sentence (in the OP) is more like answering to the question asking 'Who likes cats?' while the other one (which I showed here) is answering to 'What do you think your teacher likes?'

I wouldn't say 私は先生が猫が好きだと思います... Why not? It just sounds unnatural... Well I might say that but wouldn't write that at least.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.