What's the etymology of まほし? Considering it goes after the 未然形, is it something like む -> ま (未然形) + [欲]{ほ}し, or is it a totally different etymology altogether?

Thanks in advance.

1 Answer 1


The -(a)maposi optative auxiliary adjective is indeed connected with ほし "want," but the first part is different. It is derived from the simplification of -(a)maku [nö] posi, where:

  • -(a)m- is trivially the ordinary modal (Classical む/ん);
  • -aku is the nominal ending;
  • posi is ほし "is desirable."

In Nara Old Japanese, the construction is still attested in the non-contracted form:

栲領巾乃懸巻欲寸妹名乎 (MYS 3.285)
taku pire-nö kakë-m-aku posi-ki imo=nö na=wo
"I want to say the name of my love as if hang the scarf of mulberry"

Literally, "hanging is desirable." When the nominal form in -aku was lost, the construction was grammaticalized.

In Old Japanese, -(a)maku [nö] posi is only used with a limited series of verbs (citing Vovin): kakë- 'to hang = to mention,' kagapur- 'to obtain,' kik- 'to hear, to listen,' mor- 'to guard, to protect,' mi- 'to see, to look,' sömë- 'to dye,' ne- 'to sleep.' Since Heian, the grammaticalized descendant loses these limitations, and even in Taketori we find:

あるいはをのがゆかまほしきところへいぬ "As for some, they went towards the place they wanted to go," with yuk- "to go" not in the previous list. Probably this implies it was completely analyzed as a part of the verb's conjugation pattern.

Sources and further reading:

  • Frellesvig, Bjarke - A History of the Japanese Language - Cambridge University Press (2010), pp. 240-2
  • [Handbook of Oriental Studies. Section 5 Japan 16] Vovin, Alexander - A Descriptive and Comparative Grammar of Western Old Japanese (2 Vols.) - Brill (2020), pp. 694-8.
  • Thank you, that makes perfect sense. I actually saw 〜まく欲し many times before (it's even used as in example sentences in a few dictionaries) but never made the connection between it and まほし. Many thanks! Commented Jun 15, 2021 at 15:50
  • I'm curious, where does the no come from? This generally only connects two non-inflecting words, and ほし inflects as an adjective. I'm not aware of other cases of [NOUN][ADJ] constructions like this. Do Frellesvig and Vovin insert the no here? If so, do they explain why, and what function it has? Commented Jun 15, 2021 at 16:19
  • It is not hard to find attestations of that: 4.584 見巻之欲寸 mimaku nö posiki, 10.1913 見巻之欲 mimaku nö posiki, 20.4449 美麻久能富之伎 the most blatant, mimaku nö posiki (funnily, all with the same verb, but it is generally the most frequent to be used in this construction.) Old Japanese adjectives are much more of nouns than on the later stage, and probably this shouldn't be considered something special. Commented Jun 15, 2021 at 16:39
  • Very interesting. Digging around further, the 日本国語大辞典 entry's sense [1].[四].② gives the definition 「好悪の感情や希望・可能の対象を示す」, with a quote from MYS 11.2554 using the same mimaku nö posiki construction. Parent sense [1].[四] defines this usage of の more broadly as 「主格を示す助詞」, which would fit this [NOUN][ADJ] grammar here before が developed its "subject" sense. 古語の勉強になりました。ありがとうございます。 Commented Jun 15, 2021 at 18:03
  • A later afterthought. The great expansion of the verbs allowed with まほし between Man’yōshū and Taketori could more plausibly be explained by dialectal shift, from the speech of Nara capital to the speech of Heian region. It is likely that the Nara dialect was especially conservative in this aspect, not that suddenly lost of new verbs were allowed. Commented Mar 5, 2022 at 9:49

You must log in to answer this question.

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