Am I correct in thinking that these are ok without a na:

1) 印象的事実
2) 印象的事実
3) 客観的事実
4) 客観的事実

But these need a na?:

5) 明らか事実
6) *?明らか事実
7) 簡単事実
8) *?簡単事実

(From a quick look, Google results count seems to support this).

According to Daijirin, -teki should be making na-adjectives:


But is it the case that these -teki-adjectives (and only they) accept dropping the na? If so, why is that? Influence from the original Chinese particle?

Is there any linguistic articles on this topic?

  • 1
    Could it be the "newspaper style" where some particles are omitted? – comeauch Jan 4 '15 at 4:19
  • 印象 and 客観 are nouns. Hence 2) and 4) can be interpreted as compounds. 明らか and 簡単 are nominal adjectives. – Thomas Gross Jan 4 '15 at 6:50
  • 2
    @ThomasGross 印象 and 客観 are nouns, but they're not compounding with 事実 in 2 and 4 (it's not 客観事実、印象事実); rather, they're modified by 的 first, and 的 makes nominal adjectives out of nouns (just like English "-ive" in "object"→"objective"); cf. dictionary link. I know that some Japanese words can work as both nouns and NAs (e.g. 特別の/特別な); but if this was the case of 的-adjectives, we'd expect not only compounding but also *?印象的の事実、*?客観的の事実; however, these get no more than a handful of hits, not many more than *?明らか事実。 – melboiko Jan 4 '15 at 13:50
  • 1
    As you mention 的 is equal with -ive in English, without な looks "cool". Really. Now do not use の。It is grammatically incorrect. And also you might better consider which Kanjis fit with the fact. Ambiguous -ing kanji such as 印象的 does not fit with the "concrete" 事実。Sounds strange to us. Good luck. – Kentaro Tomono Jan 4 '15 at 17:55
  • 1
    @ThomasGross Pitch accent tells us that 2 and 4 can't be interpreted as compounds. Although in this case we can't tell because the pitch accent would be the same either way, if we pick an unaccented second word the difference becomes clear. See Martin 1975 p.762-763, where we can find the example 経済的問題. – snailboat Jan 4 '15 at 23:01

When to drop "な" depends on the phrase. 客観的事実 is a very common set phrase, but 印象的事実 is not. You have to usually say "それは印象的事実(でした)", unless you were a philosopher and ready to give 印象的事実 some definition.

的 is not special; there are several kanji which can connect two nouns and help to make longer compounds without hiragana particles.

~風 (~ style)

  • 西洋風の建築 → 西洋風建築 Western-style architecture
  • 今風の若者 → [?]今風若者

~用 (for ~)

  • 開発用のソフトウェア → 開発用ソフトウェア development software
  • 冬用のテーブル → [?]冬用テーブル

~的 (~'s, of ~)

  • 一般的な事実 → 一般的事実 general fact
  • 一般的なバナナ → [?]一般的バナナ

~型 (~ type, ~ shape)

  • 朝型の人間 → 朝型人間 morning person
  • ハート型の雲 → [?]ハート型雲

~後 (post ~), ~前 (pre ~)

  • 出生前の診断 → 出生前診断 prenatal diagnosis
  • 飲酒後のジョギング → [?]飲酒後ジョギング

Here, [?] indicates that expression is probably grammatical but not familiar at least to me. Dropping particles (の/な) will make that group of words sound as "one set phrase". Unless you really want to introduce that as an independent idea (e.g. "after-drink jogging", "heart cloud"), dropping particles would sound more or less strange.

As you can see, I think the similar omission happens in English, too. "Earth axis" is one of the common set phrases you can find in every dictionary, and saying "axis of the earth" all the time is too bothersome. But that does not mean you can freely treat "earth" as an adjective.

  • What's interesting about 的, compared to your other examples, is that 的 asks for a な and not a の, which makes the 的-word syntactically a na-adjective. Noun+noun compounds without the の is totally common, but it seems to me that few na-adjectives work without a な, compounding directly on what Japanese school grammar call the "stem" (語幹、 the part before な). Thanks for the additional information, it's deeply appreciated. – melboiko Jan 5 '15 at 8:37

Matt Treyvaud answered my question elsewhere, so I'm posting his answer here.

  • Yes, teki-words are unlike other na-adjectives in that they can modify nouns without a na. The Daijirin definition was incomplete. From the Nihon Kokugo Daijiten:

漢語について、直接、または「な」 をともなって連体修飾語として用いられるほか、形容動詞語幹として使われる。

  • Contrary to what I proposed above, teki-words seem to be able to work as nouns, after all. A closer look on the web and Twitter found examples of teki-words followed by を or subject が. Again from the Kokugo Daijiten:


  • And yes, this is due to influence from Chinese grammar, from where the morpheme was borrowed in the first place. But Japanese teki is a lot newer than I expected (de 的 isn't, after all, a Literary Chinese particle; that would be zhī 之). Turns out that teki-words are as recent as Meiji. In its early days, teki has even seen use as a general nominalizer:


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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