I read this in Pulleyblank's Outline of Classical Chinese Grammar, where he said it is a maybe.

In Old Chinese (OC), they have the following initial and rhyme groups with Baxter-Sagart reconstructions:

p + juw (*pə)

乎 h + u (*ɢˤa)

夫 b + ju (*ba)

My main concern is that 夫 fu2 has a voiced initial in OC [b] (note that it's not f that I am taking about, since f was a new development from the b group initials in late Middle Chinese), while 不 has a unvoiced one [p].

Is there any other theory, or other evidence so that this theory stands?


As @水巷孑蠻 mentioned, the well-recognized sound fusion includes 不 + 可 > 叵, which the initials differ only in aspiration. If difference in aspiration is not a problem, then I think difference in voicing is also not a problem.

But @水巷孑蠻 raised another problem, which is that if we rewrite 夫 to 不乎, then the sentence reads odd. For example, 吾歌,可夫? -> 吾歌,可不乎?

Yet I did find some cases in later documents the following examples:

不乎, 宰我以仁者必濟人於患難,故問有仁人墮井將自投下,從而出之不乎?


2 Answers 2


Is the final particle 夫 a fusion of 不乎?

very doubtful, imo

“夫” used as a particle for indicating the end of a verse, has the functions of exclamation or query; 文言文中用於句末,表示感嘆或疑問。

about the “fusion”, in the past, some “argued” that “反切” was an ancient han-chinese method, not imported from bharat

eg, in the book 夢溪筆談 卷十五 by 沈括


that the phonetic mechanism of the fusion “二聲合爲一字” is

bat1 可 hoh2 爲 叵 poh2 (b, p pronounced loosely the same, loosely)

hoh4 不 bat1 爲 盍 hap6 (both share the phoneme /a/, entering tone)

yue4 是 si6 爲 爾 yi5

yi4 已 yi5 爲 耳 yi5 😹

ji1 乎 foo4 爲 諸 jue1 (for this one, cantonese can’t show the fusion 😼)

phonetically, the notion “夫 a fusion of 不乎” is, . . . unappealing

semantically, the left and right side of the fusion should be interchangeable, eg

人心叵測 = 人心不可測

盍各言爾志 = 何不各言爾志

it’s odd to rewrite 夫 to 不乎, eg

“逝者如斯夫,不舍晝夜” —-> 逝者如斯不乎,不舍晝夜

“孔子曰:『吾歌,可夫?』” —-> 孔子曰:『吾歌,可不乎?』

have fun :)

  • phonetically, if 不 + 可 -> 叵, then I think 不+乎 -> 夫 is ok. The only problem is that the initials 不 is [p] and 夫 [b], differing in voicing. Not perfect, but given that the well-recognized pair 不[p] and 叵 [ph] differ in aspiration, I'd say the difference in voicing is also acceptable. And semantically, I do agree with your examples, but also found some later examples with 不乎, 宰我以仁者必濟人於患難,故問有仁人墮井將自投下,從而出之不乎?and 而父卒祖母後卒,當服三年不乎?
    – lilysirius
    Commented Nov 9, 2023 at 7:03


Those are pronounciations of those characters in putonghua, which are very likely to be different from how they were spoken way back then.

  • Actually it's the opposite. In Putonghua their pronunciations are very different, but in Old Chinese they were close. If you look at my question closely, you will see that I have given their initial and rhyme groups and reconstructions in OC. My question never concerns Putonghua.
    – lilysirius
    Commented Nov 10, 2023 at 3:46
  • I am not sure what "Old Chinese" refers to, as the pronounciations change over the years. At any rate, in Cantonese, which is closer to "Old Chinese" than Putonghua, 不 is annoted as bat1, 乎 as fu1 and 夫 as fu1, according to the Hong Kong Chinese University website. So it still seems unlikely. Commented Nov 10, 2023 at 10:20
  • 1
    For the start, the initial f is a new development from the b/p group in late Middle Chinese. Before that, Chinese had no such sound. Also, Cantonese is a direct descendant from Middle Chinese, so is Mandarin. They both kept some characteristics of Middle Chinese, but they are still very far from Old Chinese. Mismatched in Cantonese doesn't not say anything in terms of my question. Btw, Old Chinese is roughly Chinese before 4th century.
    – lilysirius
    Commented Nov 10, 2023 at 11:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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