There are a couple of posts suggesting that, with certain verbs, が can actually mark the direct object of a sentence instead of the subject. But it seems to me that, in these cases, we can just reinterpret the words marked by が as being proper subjects (instead of direct objects). The purpose of this post is to see if this model of Japanese is getting something wrong?

Example 1: 分かる. An example of a verb satisfying this is 分かる:

  1. 「ここが分かる」 "This is understandable." ("This" as the sentence's subject)
  2. 「ここを分かる」 "I understand this." ("This" as the sentence's direct object).

Example 2: Potential verbs. Then there are the potential cases, like:

  1. 「新聞が読める」 "Newspapers are readable." ("Newspapers" as the sentence's subject)
  2. 「新聞を読める」 "I can read newspapers.* ("Newspaper" as the sentence's direct object)

This interpretation also seems to preserve the (elsewhere cataloged) nuances of が vs. を:

  • Using を makes the sentence sound more volitional (since e.g. in the cases above, it forces the subject of the sentence to be a human being, rather than an inanimate object).
  • Using が gives the sentence an exclusionary feel (e.g. "This (as opposed to other things) is understandable" or "Newspapers (as opposed to other things) are readable.).

Is there something wrong/misleading in interpreting が this way (essentially, insisting that が never indicates direct objects)?

  • 2
    が only marks "objects" if you're thinking in English. See also this other post in response to a question about "dative subjects" and "nominative objects". Feb 14 at 4:40
  • 2
    新聞が読める is pretty neutral. は instead of が would give the sentence an exclusionary feel.
    – aguijonazo
    Feb 14 at 5:28
  • 1
    @sundowner: What if translated it as "Newspapers are readable (by me)"?
    – George
    Feb 14 at 16:26
  • 1
    This seems related too.
    – aguijonazo
    Feb 14 at 17:14
  • 1
    @George, considering the historical development of the -eru potential forms, and given the consistent grammar of using が to mark the subject, yes, I don't think there's much of a sensible case to make for が marking "objects" as Japanese -- that only works in "Japanese as viewed through an English lens" (or perhaps the lens of some other PIE-derived language). For that matter, Spanish's use of the reflexive for potential is somewhat similar, in that the "thing" that "can be [VERB]-ed" is marked as the reflexive subject. See constructions like "se habla español". Feb 14 at 17:37

2 Answers 2


Without any further context, the most reasonable interpretation of 「新聞が読める」 is that "I am able to read a news paper".

To say, "Newspapers are readable" says absolutely nothing about whether I can or cannot read a news paper. This construct in English has a somewhat passive feel to it (how different is from saying "Newpapers can be read").

But the Japanese isn't passive at all.

Is there a reason you're not happy with が marking the subject? This is how ergative languages can work.

  • 2
    Isn't parsing "新聞が読める" as "newspapers are readable" fine so as long as we admit there is an implicit "by me" at the end (or perhaps an implicit 「私は」 at the beginning of the Japanese sentence)? Also, I'm totally happy with が marking the subject :] In fact it's allowing が to mark direct objects which I find really confusing here, and what motivated me to make this post. My goal is to see if I can safely parse sentences like these as "が-subject sentences", so I can preserve the notion that "が always marks the subject".
    – George
    Feb 14 at 16:21
  • 1
    @George, the が here marks the subject of the potential verb 読める. Here, 読める is more directly translated as "to be readable", not "to be able to read". Originally, and still commonly, Japanese verbs of potential (and desiderative verb forms using ~たい) describe a characteristic of a thing. By contrast, English verbs of potential describe an ability of an actor. It's a different focus. Feb 14 at 17:47
  • 1
    @George, I think this is also why we start to see を used with potential verbs: this shifts the focus from the thing that is [VERB]-able, to the whole verb phrase being possible. 本が読める ("the book is readable [by me]", emphasis on what is readable) → 本を読める ("it is the reading of the book that is possible", emphasis on what action is possible, as a whole verb phrase). Bracketing the focus, we have(本)が読める (noun focus) and (本を読)める (verb phrase focus), Feb 14 at 17:51
  • 1
    Similar for desiderative ~たい forms: (ピザ)が食べたい ("it's the pizza that is desirable to eat [by me]", emphasis on the pizza) → (ピザを食べ)たい ("it's eating pizza that is desirable to do [by me]", emphasis on the entire verb phrase). Feb 14 at 17:51
  • 1
    @George, that sounds like a reasonable interpretation. With translation, there is always a mismatch between the source and target, so it's often a matter of figuring out where that mismatch is least disruptive (not losing important meaning) and most appropriate (producing the most fitting target text for your use case). Feb 15 at 0:19

The reason Japanese, much as Icelandic, is generally regarded as a language with “quirky cases” and that in “私にこれがわかる”, “私に” is analysed as the, quirky, dative subject and “これが” as the, quirky, nominative object is because aside from the cases they behave as subjects and objects do.

There are actually many, many examples but a basic one would be subject honorrification. “あなたにこれがお分かりです” elevates the status of the subject “あなたに”. Another case would be that clauses that use “ながら” must share the same subject, and for that purpose again “それがわかりながら、なぜ私たちはそこへアクセスしてしまうのか?” seems to treat it as the subject.

There is an endless list of similar arguments that can be made for Icelandic, Japanese, German, and many other languages with “quirky cases” why these parts of speech marked with a case one would not expect, are actually what they are in how they behave grammatically.


  • I disagree. The honorific elevates the status of the agent, not the subject. Separately, ながら constructions do not necessarily require that the following clause use the same topic or subject. See this example, first sentence of the second paragraph. Mar 14 at 19:54
  • I̵'̵m̵ ̵i̵n̵t̵e̵r̵e̵s̵t̵e̵d̵ ̵i̵n̵ ̵r̵e̵a̵d̵i̵n̵g̵ ̵y̵o̵u̵r̵ ̵l̵i̵n̵k̵e̵d̵ ̵r̵e̵f̵e̵r̵e̵n̵c̵e̵,̵ ̵b̵u̵t̵ ̵t̵h̵e̵ ̵l̵i̵n̵k̵ ̵i̵s̵ ̵b̵a̵d̵.̵ ̵ ̵C̵o̵u̵l̵d̵ ̵y̵o̵u̵ ̵d̵o̵u̵b̵l̵e̵-̵c̵h̵e̵c̵k̵ ̵a̵n̵d̵ ̵u̵p̵d̵a̵t̵e̵?̵ Never mind, fixed it! Mar 14 at 19:55
  • FWIW, the link appears to be the same paper discussed in this other post. Mar 14 at 20:00
  • @EiríkrÚtlendi Well, in passive constructions, subject honorrification still raises the status of the subject rather than the agent. There are of course many ergative-verb pairs where the subject in the unaccusative one is the patient, rather than the agent, and still has it's status raised with subject honorification. As for the example sentence, I'm not sure why “厳格なる機能” can't be analysed as the subject of the clause that ends in “わかりながら”
    – Zorf
    Mar 15 at 14:42
  • The clause ending in 分かりながら has an explicitly stated object, and that is not 厳格なる機能...? Separately, are you arguing that verbs like ある are transitive? That appears to be Shibatani's argument (as laid out on page 799 of the paper referenced by Ellen Woolford's 2000 paper on the Rutgers server, as linked in your post). I cannot agree with that: ある is only "transitive" after translation into idiomatic English using the transitive verb "have". Likewise for his other examples. Mar 15 at 22:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .