My recent article about Kanaya’s theory of not needing a subject in Japanese was somewhat controversial and I received a few good questions, that I would like to publish here as well (the original questions are from a Facebook group)
It isn’t clear to me how Kanaya or Mikami’s theory is any better than mainstream linguistic ideas of Japanese grammar and your essay doesn’t address this. (Is there any difference–in practice–between Kanaya’s framework where there is no subject at all and a subject that is implicitly there (but perhaps unspecified or unknown to the listener?)).
I didn’t intend to compare Kanaya and Mikami’s theories with the current Idea of Japanese grammar, that’s why I didn’t address it. I wanted to point out the major difference of the current Hashimoto Shinkichi’s theories on which the current Japanese education is based for teaching Kokugo. That is, as opposed to presuming that there is a hidden subject, the theory argues that there isn’t one to begin with. The practical impact of that as I see, Learners who still can’t figure out how to properly use not only particles like はandが but also other particles like 「も」「すら」「さえ」「こそ」can have a chance to understand things as it really is, Instead of trying to follow inaccurate rules which in many cases are confusing. I will elaborate on this in a separate article about「は」and「が」
“Kanaya…his definition of a complete sentence was, if you can form the negative and past tense by slightly altering it, then it stands as sentence.”
What was his justification for this set of rules? Most linguists rely on native speaker intuitions for deciding what constitutes a sentence (which is not to say this is without complications). Or his rule was devised to explain the kinds of sentences that native speakers accept as grammatical? Are there “near misses” that illustrate how some sentences are not grammatical (e.g. something that can be made negative but not past tense)?
Unfortunately I returned the book to the library, so I can’t check if he justified this set of rules, however, what I do know is that his book only gained popularity because Japanese natives can relate to his theory, and yeah, even though I’m not a native speaker, I didn’t question his definition of a complete sentence, it instantly made sense to me. and you’re very unlikely to find “near misses” because conjugation is very systematic in Japanese with little to no irregularities, if you can make the negative of a word you Can get the past and the other forms as well. but if you find such misses please let me know.
You write that 「白い。」is a complete sentence is Japanese. Is this fundamentally different than saying “White.” in English?
「白い」is an adjective phrase, normally a complete sentence in English needs more elements, but according to Kanaya’s theory, In Japanese the predicate alone can form a complete meaningful sentence.「白い」is not any different from the other example of「楽しい」which is a「形容詞文」, on the other hand, the English ”White” is just a single word and doesn’t form a sentence. The possible equivalent sentence would be” it’s white”.
Why can’t が be the subject marker and は the topic marker? If we assume there are no subjects in Japanese, it doesn’t make sense to have a subject marker. But it isn’t clear to me that Japanese doesn’t have subjects.
First, I’d like to point out that Semantics-wise the subject does exist, although grammar-wise (Syntactically) unlike English the verb doesn’t need to agree with the subject person and number. The concept of “subject” is different. Kanaya also mentioned in his book that Japanese is a pro-drop language, but didn’t elaborate on that topic. But from my understanding, Generally speaking, European languages with rich verbal agreement patterns allow for phonetically empty subjects in finite clauses ‘pro’ – Spanish, Italian etc. Those with impoverished verbal agreement, do not – English, French, German.
Some languages (Japanese) have neither subject nor object agreement regularly allow argument omission/a pro-drop patterning. How can this patterning be reconciled with agreement-licensed pro-drop?
(This might not be accurate, so you might need to refer back to the original references)
Huang 1984: pro-drop is licensed to occur either where a language has full agreement, or where a language has no agreement, but not where a language has impoverished partial agreement.
Speas (1994, 2004): The Licensing of Agreement
- Proposal: poor agreement, rather than pro, needs to be licensed.
- If I0 has only partially-specified phi-features, these must be provided by an element in SpecIP.
- Pro lacks phi-features, and so cannot license a partially-specified I0
- Languages with poor agreement therefore are not pro-drop
- Languages with rich agreement do not require an element in SpecIP to license this agreement, and so are pro-drop
- If there is NO agreement in I0 (Chinese, Japanese etc), there is no agreement to be licensed, and so pro can occur in SpecIP
Pro occurs either with rich agreement or no agreement.
Tomioka (2003): He proposes that Japanese pro is a null NP whose descriptive content is pragmatically retrieved: the same semantic tools that are used to interpret full NPs are used to interpret pro. Tomioka suggests that what underlies discourse pro-drop is the fact that languages (almost) universally allow phonologically null NP anaphora (also known as N’ or NP ellipsis). In a language that lacks determiners (such as Japanese), this operation will give rise to phonologically unrealized arguments. In languages in which DPs are necessarily projected, a remnant D will always show up and so this process will never give rise to a silent argument.
In other words, according to this analysis, zero pronoun is not pro but ellipsis cite.
Of course each analysis has its problems, but the point is each analysis interprets zero pronoun as a different type of empty category.
Huang: the zero pronoun of languages like Italian is pro, and the zero pronoun of languages like Japanese is a topic operator variable.
Speas: In languages like Italian and Japanese, zero pronoun is pro
Tomioka: In languages like Italian and Japanese, zero pronoun occurs due to NP ellipsis.
What I’m trying to say is that even though Japanese is considered pro-drop language, or NSL, it’s really theory-dependent.
“So what about 「が」, for the above example 「鼻が長い」the 「が」after「鼻」is functioning as a modifier to add more information about its preceding word. And again this particle is not to mark the subject.”
“Nose” used in similar constructions in other languages would function as the subject, so to say that “it is ‘extra’ information but not the subject only in Japanese” is a little suspicious to me.
The difference between はand が is very controversial and needs a lot of explaining, but the least I can tell you, is that both particles have more than one function.
|Her husband is German||I miss my mother|
|The first floor of this building is a restaurant||I want to eat something sweet|
|Italy is the home of pizza||I can speak English|
You can see from the example sentences above sometimesは has possessive functions as in 「あの人はご主人がドイツ人です」it can be rephrased as「あの人のご主人はドイツ人です」this is also applies for 「象は鼻が長い」although it gives a different nuance. As for 「が」the 3 examples on the right all represent usages where it can’t be considered a subject marker yet all the above examples are valid and used in every-day situations.
I suppose this answer is not enough so I will write a separate article later about the different usages of 「は」and「が」
Also, regarding “は” and “が”–the two people, who were cited as believing that one of these is the subject marker, both were born in the 1800s and have long been dead (much before the emergence of generative grammar). What do Kanaya’s contemporaries think of “は” and “が”?
I will need some time to research this well, and I will write an article about it soon.
“Another example is words like 「我が国」「我が家」The 「が」in such sentences is modifying the preceding word and adding more details or information about it.”
I’m not familiar with these constructions. Is this the same “が” as the “subject marker” が? Can I use this noun in a sentence like:
我が国がいいです。where the second が would actually be the subject marker? Can I create arbitrary nouns with が (in a similar pattern as 我が国), or does this が only appear in a limited set of words?
Originally it is the same particle, however its usage have developed over the years, even this sample sentence has come to hold a different nuance than it used to, generally, phrases using 我が is still being used, however, it implies a level of endearment regarding what comes after it, for example 「我が国」 implies feeling proud of one’s country, other examples of this is 「我が意」「我が息子」「我が家」「我が社」「わがまま」and just today I was speaking with a Japanese friend about Kanji and he told me「我が国語は複雑怪奇也」other examples of the particle is of course the Japanese national anthem’s first words「君が代」all of which imply possessive features of「が」so it’s like saying「我の家」「私（私達）の家」of course rephrasing it changes the nuance.
Here is another example「夢を叶わんがために、努力している」here the form of 「未然形＋ん・がため」is a now a rigid expression, and is used mostly in writing, it’s also one of the grammar points in JLPT N1 test. It means “in order to” or “for the purpose of” also implying strong emotion to do something and having the will to achieve it.
So, yes you can say「我が国がいいです」But it doesn’t sound so natural, people usually say「我が家が一番」which is practically the same construction.
I know my answers are not that satisfying, there are really a lot of theories on this topic, which is also one of the reason’s that make it so confusing. and I would need to write a book to explain this thoroughly. But I would like to write more on this later so check my blog later
- Huang, C.-T. James (1984) “On the distribution and reference of empty pronouns,” Linguistic Inquiry 15:531-574.
- Speas, Margaret (1994) “Null argument in a theory of economy of projection,” University of Massachusetts occasional papers in linguistics. 179-208.
- Tomioka, Satoshi (2003) “The semantics of Japanese null pronouns and its cross-linguistic implications,” The interfaces: Deriving and interpreting omitted structures. 321-339. John Benjamins.