On this site, when asking about でしょう・だろう, as well as ましょう etc, I have been told that it is a combination of the verb's 未然形 and the 推量 particle "う". However, I have never been able to find anything on it at all, I was wondering if anyone has read about it, seen any other examples of it and could link me to those.

My question is, is this う particle real?

The reason I ask, is because it would be easy to understand 泳ごう for example as 泳が + う which would mean that the action of swimming hasn't happened yet, but you are making a guess that it will.

BONUS: Could this う particle be related to the fact that all verbs end in う?

  • 2
    – Robin
    Commented Mar 12, 2014 at 9:57
  • 1
    No, that particle does not exist because it is an auxiliary verb, not a particle.
    – user4032
    Commented Mar 12, 2014 at 10:21
  • Hey, why the downvote? TokyoNagoya that sounds intriguing, could you elaborate?
    – Nathan
    Commented Mar 12, 2014 at 10:28
  • @Nathan Last time you brought this up, I linked to these three definitions. This question is interesting, though, because 助動詞 inflect but う does not, and because (in my opinion) it's hard to come up with a good argument against considering the entire suffix a single morpheme in modern Japanese.
    – user1478
    Commented Mar 12, 2014 at 11:03

1 Answer 1


This is not exactly an answer to your question. originated from む, I think there should be no doubt. So 泳がむ→泳がう→泳ごう, the first form is used in old Japanese, the second form is used before World War II.

I think the problem is actually not whether う is a real particle, suffix or auxiliary verb, but whether 未然形 itself exists, or if the name is reasonable.

Personally, I believe the 未然形 cannot exist as a standalone form. In fact, the criteria used to classify the so-called “活用形” (namely 未然形, 連用形, 終止形, 連体形, 已然形, 命令形) are controversial.

連用形, 終止形, 連体形 are named after their functions. 未然形, 已然形 and 命令形 are named after their shapes. The problem is that verbs often have more than one 連用形/終止形/連体形's and their 未然形/已然形/命令形/連用形 may be the same.

You can see the chaos in this entry.

り [助動][ら|り|り|る|れ|れ]《四段・サ変動詞の連用形に「あり」の付いた語、例えば「行きあり」「しあり」の音変化形「行けり」「せり」の「り」から》四段動詞の已然形、サ変動詞の未然形に付く。ただし、上代では四段動詞には命令形に付く。

The rules in fact are unbelievably simple: 連用形(-i)+あり(-ari) becomes (-eri), that's all.

The only situations involving 未然形 that I can recall are:


It seems that ク語法、り(助動詞) all look like the result of assimilation of two syllables. ず(否定)、ぬ(否定)、ない(否定) share the same origin. The form used before る、す、まい, べし, etc. is not stable, there are often exceptions.

In fact, I have never seen 未然形 used without a suffix, nor is it separable. That suggests that 未然形 along with the suffix might be viewed as an inflected form, or a derived verb.

But there are many adjectives ending with the vowel /a/ or /ashi/ sharing the same stem with a verb, which makes me think that the vowel /a/ might have been widely used to express an kind of appearance. (maybe = 様{さま})

あか (said to be related to 明け{あけ} )
暖かい{あたたかい} -ai (said to be related to 熱い{あつい} )
明らか -aka
いやらしい -rashi
願わしい -ashii
望ましい -ashii

It seems that /a/ is often used to fill the gap between two consonants. As modern Japanese people tend to perceive the unstressed, reduce vowel (/ə/ schwa ) as /a/, is there any possibility that these /a/s actually reflected the old neutral vowel /ə/?

  • 1
    Three comments (^^;): -Old Japanese *ə becomes Middle Japanese /o/ in all cases, so I highly doubt this -a- is from *ə - this kind of phoneme reinterpretation only really happens when loaning, not within a language. -I feel like a lot of uses of the 未然形 do have a common semantic domain, namely, things that are not completely true. -あか and あけ are related, the second one is *aka-e- with the transitivity flipper / inchoative -e-; and I don't know if -raka has anything to do with the 未然形.
    – Sjiveru
    Commented Mar 13, 2014 at 2:28
  • @Sjiveru, Thanks for your comment. Isn't the transition /ə/→/o/ just a hypothesis. We can go even further: Old Japanese did have /VV/ structures, with /VəV/ as a variant, but when it's recorded in kanji, which does not allow /VV/ structure, /Va.V-/ was chosen to represent it. I just feel /a/ is most commonly used for filling the gap between consonants. I found a interesting article, which suggests 曽我氏{そがし} might be actually 孫氏{そんし}. lol
    – Yang Muye
    Commented Mar 13, 2014 at 3:23
  • 1
    I'm... not sure I understand what you're saying. *ə > /o/ is a hypothesis, but one that seems to be pretty darned well borne out by the data (see Marc Miyake's reconstruction for the whole reasoning). By 'V' here, do you mean a consonant? I don't know that there's any evidence at all for consonant-consonant sequences in Old Japanese - if there were single consonants, the metre in the Man'youshuu would be completely screwed up. (I also truly doubt people would change pronunciation to fit spelling.) You can totally write consonants with kanji, though - Old Korean uses 乙 for /l/, for example.
    – Sjiveru
    Commented Mar 14, 2014 at 18:20
  • あか is likely related to 赤・明け・開く. 暖かい is older 暖か (a 形容動詞) + し to make a regular 形容詞. 暖か appears to be あたた (probably from あつ, but reduplicated たた is unclear) + adjectivizing suffix か. 明らか is 明き + ら + か, with the latter two both adjectivizing suffixes forming 形容動詞 and meaning "XX kind of state". らしい is from older らし, possibly that same ら + し to form a 形容詞. For the last two examples, many modern しい adjectives derived from 未然形 verbs + し. Other examples include 疑{うたが}う > 疑{うたが}わしい, 思{おも}う > 思{おも}わしい, 好{この}む > 好{この}ましい. More at Wiktionary. Commented May 20, 2014 at 8:31

You must log in to answer this question.

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