i was reading something about adjectives on https://www.imabi.net/ and as i was reading, i came across this sentence, the なる-連体形 may follow nominals whereas the かる・しかる-連体形 cannot. is there something i am missing? i thought 連体形 is there to follow the nouns right? why is that? is my interpretation of this whole text wrong and i am thinking about it badly?

  • 3
    This is only a guess but I'm thinking the ~かる type is used before a particle (you didn't need a nominalizer like の in Classical Japanese) while you would use ~き for noun-modifying. I am definitely not a Classical Japanese expert, though; this is only a guess, which is why it's a comment.
    – Angelos
    Jan 1 at 20:11
  • 1
    i've gone through some examples and stuff, and yeah it only seems to attach to suffixes.
    – sieman
    Jan 3 at 14:48

2 Answers 2


Clarifying the context of the question

The particular Imabi page referenced in the question appears to be this one: "Classical Adjectives II".

The specific paragraph at issue, in the 「ナリ型形容動詞」 towards the top of the web page:

Similar to 形容詞, the に-連用形 is used with conjunctive particles and may be used adverbially. Since all but one base is a ラ-変 conjugate, the なる-連体形 may follow nominals whereas the かる・しかる-連体形 cannot.

Problems in the Imabi text

This description is confused.

The 連体形【れんたいけい】 or "adnominal form" of any verb or adjective can follow (that is, "come after") any appropriate noun or phrase being used as a noun (a "nominal"). So the description above is off the mark in this regard.

I suspect what they intended is that the 連体形【れんたいけい】 of this specific ~かる conjugation ending for adjectives cannot be followed by nominals (that is, "nominals can't come after this").

But why is this? Let's dive in.

Development of the ~かり conjugation paradigm

This doesn't seem to be discussed in the Imabi content I've seen so far.

The various ~かり・~かる etc. endings for -i adjectives all come from fusion with the regular ~く ending, plus the Classical copular ("to be") verb あり.

Let's look at this with the Classical adjective 高【たか】し ("high; tall"), which became modern 高【たか】い.

Conjugation Form Adjective Only Adj + あり Broken Out Adj + あり Fused
Irrealis (not happened yet)
(*Note 1)
[高]{たか}く + あら 高【たか】から
Continuative / Adverbial (modifies verb or verbal phrase)
[高]{たか}く [高]{たか}く + あり 高【たか】かり
Terminal / Predicative (ends a sentence)
[高]{たか}し [高]{たか}く + あり 高【たか】かり
Adnominal / Attributive (modifies noun or nominal phrase)
[高]{たか}き [高]{たか}く + ある 高【たか】かる
Realis / Subjunctive (as if it has happened)
(*Note 2)
[高]{たか}く + あれ
(*Note 2)
(*Note 2)
Imperative / Command
- (*Note 3) [高]{たか}く + あれ [高]{たか}かれ
  • Note 1: By some analyses, -i adjectives don't have any 未然形【みぜんけい】 or irrealis form, as there is no action or change of state inherent in the meaning of this class of words, and thus no means for this non-action to "not happen yet". Most materials that I've seen still include this form, since this is the classic explanation for negatives. An alternative view is that negatives for -i adjectives are simply the adverbial ~く form + the negative auxiliary ない. The fact that you can validly insert certain particles between the ~く and the ない appears to bolster this argument, in constructions such as 高【たか】く[は]{●}ない or 難【むずか】しく[も]{●}ない
  • Note 2: The ~けれ ending for -i adjectives is often explained as the fusion of ~く + あれ (such as in the Japanese Wikipedia here, last paragraph of that section). However, this makes no sense phonologically: in all other cases of ~く + あ~ fusion, this results in ~か~, not ~け~. And, indeed, there was a ~かれ realis ending for adjectives during the Edo period, used in apparent free variation with ~けれ (although this ~かれ for the realis appears to be obsolete in modern mainstream usage).
    I suspect that ~けれ is instead a development from the older ~け realis ending found in Old Japanese. The ~けれ adjective ending is already attestable in the 後撰【ごせん】和歌【わか】集【しゅう】 dated to the mid-900s.
  • Note 3: There is no imperative form for base -i adjectives without the addition of a verb. Much like in Note 1 above, without a verb, there is no action to command someone to do.

The forms fused with あり ("to be") are mostly obsolete in modern usage. There are a couple exceptions, where these forms persist:

  • The imperative (or possibly the Edo-period realis?) is still used in modern Japanese in specific phrases, such as 多【おお】かれ少【すく】なかれ ("more or less"), basically parsing out as "be it more, be it less".
  • The ~かった past-tense ending for modern -i adjectives is from this same ~く + あり fusion, using あった in the past tense.


Historically, the forms fused with あり and its conjugations were only used in those cases where the bare adjective itself wouldn't fit the grammar -- such as when the adjective was followed in turn by a subsidiary verb like べし ("should") or まい ("probably won't"), which only attach to verbs.

Since the adnominal or attributive form of the base adjective already does the job of modifying a nominal, there was never any real need for the fused adnominal form ~かる when modifying a nominal. This is probably why ~かる + [NOMINAL] constructions are so rare.


形容動詞(adjectival noun) is divided into two types, named ナリ活用 and タリ活用. There is only one 連体形 for each, ~なる(in ナリ活用) and ~たる(in タリ活用).

形容詞(adjective) is divided into two types, named ク活用 and シク活用. But there are two 連体形 for each, ~き・~かる (in ク活用) and ~しき・~しかる(in シク活用). ~き and ~しき is mainly followed by the nouns. And ~かる and ~しかる is mainly followed by the auxiliary verbs. However, the usage followed by the noun does not seem to be completely unused. Actually, I find an expression ”隙多かる板屋”(the house with many gaps) in 源氏物語, but I don't find other example.

You must log in to answer this question.

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