1

I've been searching for the etymology of ない for a while now, and can't seem to find a clear answer. To be clear, this is the 助動詞 (auxiliary verb) ない, not the adjective 無い (Classical Japanese 無し). ない seems to somehow derive from the archaic 助動詞 (later ず), but how?

One theory I've found says that ない is derived from the Eastern Japanese negative 助動詞 なふ, used in parallel with Kansai . This checks out, given that ない seems to have originated mostly in Eastern Japanese. How would this derivation have happened—is なふ > 連用形 なひ > ない feasible? Is this the likely etymology of ない?

In any case, what is the etymology of なふ? One source says it's (未然形 of ぬ), plus a 助動詞 .

I realize that this is starting to sound less like a question and more like a theory. My questions are:

  • Is the descendancy path I've laid out for ない (ぬ > な > なふ > なひ > ない) likely?
  • Are there any other likely etymologies for ない?
  • Is ない related to ?
2
  • This question is similar to: where does ~ない come from. If you believe it’s different, please edit the question, make it clear how it’s different and/or how the answers on that question are not helpful for your problem. Commented Jul 6 at 9:07
  • @EiríkrÚtlendi That question only gives one possible derivational origin of ない, and explains why it's unlikely ("there are problems with this theory"). I'm looking for if there are any more likely etymologies, or if the one I've set out is the most likely one. I'm also wondering if ない is related to ぬ.
    – ookap
    Commented Jul 8 at 2:36

1 Answer 1

2

As I mentioned in the earlier Q&A thread about this, there are problems with any attempt to directly link なふ and modern ない.

Phonlogy: the sounds and sound shifts

One specific aspect is phonology, in terms of the sound shapes of the different conjugation forms.

If we see all of the forms laid out, the case against any direct connection becomes a bit clearer. So let's compare the conjugations, looking at the older classical forms. The headers link through to the 日本国語大辞典【にほんこくごだいじてん】 ("NKD") entries at Kotobank, and the romanizations reflect the so-called ハ行転呼, or "h-row sound-shift" that occurred for medial (mid-word) "h" kana in the late Heian period. This is where the Old Japanese (and probably earlier) /p/ sounds shifted first to technically /ɸ/, a bilabial kind of "f", and then shifted further in medial positions to /β/, basically a "w" sound.

Form For ~なふ For ~ない
未然形【みぜんけい】
Irrealis / Imperfective
~なは
-nawa
-
連用形【れんようけい】
Continuative / Adverbial
- ~なく*
-naku*
終止形【しゅうしけい】
Conclusive / Terminal / Predicative
~なふ
-nawu
~ない
-nai
連体形【れんたいけい】
Attributive
~なへ
-nawe
~ない
-nai
已然形【いぜんけい】
Realis / Subjunctive
~なへ
-nawe
~なけれ*
-nakere*
命令形【めいれいけい】
Imperative / Command
- -

(Notes: I omit the irrealis form ~なから included in some references, as this is a fusion or contraction of adverbial ~なく and the irrealis stem あら of copular / existential verb あり・ある. Forms marked with an asterisk are later developments, described further below.)

As we can see here, the ~なふ negation suffix never had any continuative form, so there never was any ~なひ (-nawi) that could shift to ~ない (-nai).

Theories on the derivation of ~なふ

The NKD entry for ~なふ mentions in the 「なうの補助注記」 section that one theory on the origins of this suffix is that it might be from the irrealis form ~な of negation suffix ~ず, plus the iterative / repetitive / stative auxiliary verb suffix ~ふ.

The problem with that is conjugation forms.

Form For ~なふ For ~ふ
未然形【みぜんけい】
Irrealis / Imperfective
~なは
-nawa
~は
-wa
連用形【れんようけい】
Continuative / Adverbial
- ~ひ
-wi
終止形【しゅうしけい】
Conclusive / Terminal / Predicative
~なふ
-nawu
~ふ
-wu
連体形【れんたいけい】
Attributive
~なへ
-nawe
~ふ
-wu
已然形【いぜんけい】
Realis / Subjunctive
~なへ
-nawe
~へ
-we
命令形【めいれいけい】
Imperative / Command
- ~へ
-we

So iterative / repetitive / stative ~ふ has attested instances of both the continuative and the imperative, which are mising for ~なふ. In addition, iterative ~ふ has the attributive form of ~ふ, while the corresponding attributive for ~なふ is ~なへ.

We might be able to explain the missing forms for ~なふ as simply cases where semantically (meaning-wise), it just didn't make sense for there to be any adverbial or imperative forms. After all, even modern negation suffix ~ない has no imperative form.

However, the mismatch in attributive forms for ~なふ and ~ふ is harder to explain, and this suggests that the ~ふ in ~なふ is not the same thing as the iterative / repetitive / stative ~ふ auxiliary suffix.

Theories on the derivation of ~ない

The NKD entry for ~ない mentions in the 「ないの語誌」 section that one theory on the origins of this suffix is that it might be from Old Japanese ~なふ. However, as also noted in that same section, ~なふ disappears from written records while the language was still in the Old Japanese stage (basically, still in the Nara period), while ~ない as a suffix doesn't show up until the late Muromachi period — that's a gap of nearly 600 years.

When ~ない does appear, it is first attested just in the already-modernized form ~ない, where the conclusive / terminal / predicative form and the attributive form have collapsed into the single ~い ending (rather than the older conclusive ~し and attributive ~き). Its earliest attestations are just of the ~ない conjugation — we don't see adverbial ~なく or subjunctive ~なけれ until later in the Edo period. The 1603 Nippo Jisho ("Japanese-Portuguese Dictionary") records the term as Nai and notes that it is defective (doesn't have a regular full conjugation; see here, on the hit for page 174, right-hand column about 2/3ds down). Furthermore, once additional conjugation forms appear, the conjugation paradigm was a bit irregular compared to other ~い adjectives, such as a past tense of ~なんだ instead of modern ~なかった, or a hypothetical of ~ないければ isntead of modern ~なければ.

By way of comparison, this separate NKD entry also shows us that standalone ない as an adjective is attested all the way back at the earliest stages of written Japanese, since at least 712. This is attested with the normal conjugation forms expected for any ~い adjective (the so-called ク活用), without the irregularities seen with suffix ~ない.

Conclusions

The modern negation suffix ~ない probably isn't related to the Eastern Old Japanese negation suffix ~なふ, at least not as a direct descendant.

  • The sizable time gap between the last attestations of ~なふ and the first attestations of ~ない is too long.
  • The differences in conjugation forms are difficult to reconcile.
  • After ~なふ went extinct, negation suffix ~ず and related forms appear instead as a replacement.

→ My personal sense is that negation suffix ~ぬ (originally the attributive form of ~ず but already being used as the conclusive / terminal / predicative by the Edo period) and standalone adjective ない basically merged. Adjective ない is the basis of negation for adjectives (more clearly visible by how we can insert a particle between the adjective and negation, such as たかく[も]{●}ない or やりたく[は]{●}ない), and I suspect that this helped motivate the use of ない for verbs as well. The same initial /n/ sound in both ない and older negation suffix ~ぬ likely also helped.

2
  • 1
    Thank you so much for the well-thought-out, informative answer! One question—by "standalone adjective ない" are you referring to the adjective that has also been spelled 無い, or is this a separate adjective ない?
    – ookap
    Commented Jul 9 at 7:49
  • @ookap, ya, ない as in "there isn't", also attested with the spellings 無い・莫い・勿い・毋い・无い・亡い (that last one, especially in reference to a person "becoming not", as in "dying"). Commented Jul 9 at 16:23

You must log in to answer this question.

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