SuttaCentral
My Gakken dictionary says:
Early Middle Chinese (pre-Tang): 霜 sïang; 乡 hɪang
Middle Chinese (Tang): 霜 shïang; 乡 hɪang
Yuan dynasty (post-Song): 霜 shuang; 乡 hiang
It looks like the i’s were pronouncing a little differently in middle Chinese, but Cantonese is closer to the older readings in some ways, it just has a different vowel.
I have also read about this. It is because many Tibeto-Burman languages can be reconstructed as multisyllabic iirc. Chinese, Old Chinese, seems to have become substantially more “uni-syllabic” than other related languages at its time, like it was a trendsetter. Old Chinese reconstructions differ in how much they think final vowels and second syllables are retained or dropped. I’ll link a paper. I have two uncited claims in the forum I have promised to cite so I will get to that shortly.
When we talk about this ancient a stage of the language, when it is multisyllabic and inflected with verb conjugations and possibly noun declensions, we’re arguably talking about “various (very) ancient Sinitic languages” moreso than even “Old Chinese.” At that stage, it is closer to proto-Sinitic.
I’m not sure how to appraise this paper. It has examples of what we were talking about, but it has too much of a conspiratorial “we’re busting this wide open” tone for me and really makes me suspicious of some of its claims.
On page 185, it introduces a “well-known Modern Chinese example,” 甭 (beng) being previous pronounced as “buyong” (不用). On page 181 it talks about
The heretic idea – then and still today – that Old Chinese may not have been exclusively monosyllabic, despite the fact that all modern Sinitic languages have overwhelmingly monosyllabic morphemes, was probably first articulated in Chinese in an essay by the “national essence” (guocui 國粹) movement philologist and philosopher Zhang Binglin 章炳麟, entitled “A theory that one character corresponds to doubled sounds” (Yi zi chong yin shuo 一字重音說).
There is a lot about how apparently a lot of people during the 1600s used to think that the Chinese and Egyptians were interrelated peoples because they both used logograms… no joke. You can read about that on p. 161.
This paper is a bit better:
https://academicworks.cuny.edu/cgi/viewcontent.cgi?article=4705&context=gc_etds
Table 2.3 is called “Complex onset structure in OC,” OC being “Old Chinese.” It is on p. 23.
An “onset structure” refers to the consonant cluster opening the OC word. Table 2.4 two pages down is a bit clearer:
A “sesquisyllable” is a “syllable and a half” indicating what is likely a truncated section of the word. The consonant clusters take time themselves to pronounce and preserve timing units from previous stages of the language. We will notice the first column of the OC reconstruction has s-njo.
The idea seems to be that at one point the sesquisyllables were disyllables or even more phonemically complex. In the above reconstruction video from YouTube, the speaker struggled with the consosnant clusters so much that sometimes a word would end up sounding like it had four syllables. A more fluent pronunciation of the reconstruction wouldn’t get so caught up on the clusters, but many words would be sesquisyllabic, not monosyllabic like we think of modern Chinese.
Thanks. You are a scholar!
I need to go back and see where I read it. Maybe a book by Sagart (I think I read the Chinese translation).