Cantonese is both very cool and very old
« previous post |
See also this paper on "Links Between Cantonese and Ancient Middle Chinese".
I asked Don Snow, a specialist on Cantonese language, if the video and the paper are "cogent, well-done, convincing?".
His reply:
Basically, I would say yes. Obviously the video is designed to be as entertaining as possible – and I might quibble about a few things – but I wouldn't have any trouble with the general thrust and even the details.
*VHM: They're by the same author.
Cantonese is undeniably older than Mandarin, but we should not close our eyes to the longevity of other Sinitic languages, which themselves cannot be demonstrated to be primitive and pristine, inasmuch as they possess substratal and areal features, plus borrowing in various directions.
As for what the oldest / earliest Sinitic language looked / sounded like, we'll have a much better idea when Jerry Norman's reconstruction, the compilation and editing of which have just been fundamentally completed, is made available to the public at large. It "is the first one based entirely on the evidence of living forms of Chinese [i.e., Sinitic] rather than philological sources." (David Branner)
Selected readings
- "Mutual Intelligibility of Sinitic Languages" (3/6/09) — or not
- "Cantonese and Mandarin are two different languages" (9/25/15)
- "American diplomat in Hong Kong reciting a Tang poem in Cantonese" (3/21/25)
- "The Future of Cantonese" (5/27/15) — guest post by Robert S. Bauer
- Donald Snow, Cantonese as Written Language: The Growth of a Written Chinese Vernacular (Hong Kong: Hong Kong University Press, 2003)
- "Of chains and Old Sinitic reconstructions" (1/27/21)
[Thanks to Alan Kennedy]
Lasius said,
April 1, 2025 @ 10:43 am
How can any modern language be younger than another, unless it was created ex nihilo like sign languages or conlangs? Don't Cantonese and Mandarin ultimately descend from the same progenitor? If so, would that not mean that they are of the same age? You could of course argue that for example phonology may be more conservative in one branch, but that doesn't mean that one is older than the other, right?
Benjamin E. Orsatti said,
April 1, 2025 @ 11:32 am
I look at it this way — Icelandic is "older" than English. Why? Since the Eddas were penned, Swedish, Norwegian, Danish, and English all lost their case systems, and from about 1066 to 1666, English was flooded with Latinisms, Hellenisms, and a good portion of the Anglo-Saxon substrate was replaced by, well, French. My Icelandic college roommate always said that he could read the Eddas and even Beowulf without any special training just by virtue of being an L1 Islandophone.
So, in a very practical sense, Icelandic is older than English.
Jonathan Smith said,
April 1, 2025 @ 11:46 am
That in general terms Yue/Cantonese dialects are more conservative with respect to earlier tones and syllable structure (esp. consonant codas) than are Mandarin dialects is of course duh. BUT just from the tones discussion at 8:00ish forward:
* "Cantonese has three level tones, high [=1], medium [=3], and low [=6], all of which could be placed in the Level [= MC 平] tone category"
>> no, of these only Cant. 1 reflects earlier Level 平 tone; mid-level 3 and low-level 6 reflect Departing 去 tone
>> not a point as when Tang poems are read in Cant., the Level tone includes Cant. low-falling tone [=4] as well
* "the Rising [= MC 上] tones [of earlier Chinese] in Mandarin […] correspond to the third tone [=3]"
>> and to Mand. 4, where in fact are found most cognates of Cant. low rising tone [=5] words
* "the Falling tone [= MC 去] still corresponds to falling tones in both dialects [= Cant. 4, Mand. 4]"
>> no, the earlier Falling/Departing tone is continued by a falling tone [=4] in Mand. but by mid- and low-level tones [= 3, 6] in Cant.
* 時 in the cited couplet ”goes against the tonal structure in both Mandarin and Cantonese… this may be an example of the poet breaking the rules"
>> no, this was a (Low) Level tone word whose reflexes include Mand. shi2 and Cant. si4, as expected
among numerous other mistakes esp. re: grammar. Oh and the Sui dynasty of 1601 CE. But noted that #LL-Eastasia finds this info accurate and shares it! Longlive Youtube, GoogleAI, Wiktionary, TikTok, etc… where else would we get our material.
豊川祥子 said,
April 1, 2025 @ 4:05 pm
Uncritically reposting the kind of "Cantonese is the original ancient Chinese language that 秦始皇 spoke" slop (from Falungong people no less!) that your auntie posts in the family WhatsApp to nurse the (very valid) pain of ongoing language shift in China would be embarrassing for anyone with a modicum of linguistic training let alone someone with a multi-decade career studying Chinese languages.
A synchronic 9 tone analysis of modern Cantonese phonology is pretty clearly unparsimonious compared to a 6 (or 7) tone one, but it at least has some diachronic truth and utility as a tool for comparison across modern Yue varieties. What is completely unfounded and should have been setting off alarm bells immediately is the claim that this is also the exact same tonal system of the variety of Chinese described in the Qieyun (varieties, really, given that what it recorded was likely more of an idealized compromise between already divergent systems of pronunciation). This is just, flat out wrong. Are we supposed to believe that MC already had both the voicing split and the Yue voiceless checked split?
The example of 雞公 as an archaicism is also pretty bizarre. While there is very real ongoing debate as to whether the sort of "original" syntactic profile of Old Chinese patterned more with typically OV or VO languages (presupposing that this is some sort of pristine blank slate free of any contact induced change), head-initial NP structures are one area that seem much more plausibly attributed to contact with typical head-initial languages of MSEA than an archaicism.[1]
[1] Szeto, Pui Yiu. (2019). Typological variation across Sinitic languages: Contact and convergence. 10.13140/RG.2.2.25385.75362.
Victor Mair said,
April 1, 2025 @ 4:25 pm
Has it not been severely criticized?
That's what Language Log is here for.
It's a blog (not an academic journal, though it does have that function too), and I think it does its job quite well. Otherwise, I wouldn't spend half my life keeping it going.
Language Log has a very broad circulation among a wide variety of readers
Thanks for your contribution.
Terry K. said,
April 1, 2025 @ 7:52 pm
I think, with Mandarin and Arabic, if we are talking about the standard forms, we can date their creation and say they are young languages. Not created ex nihilo like some sign languages, but, still, something we can mark as a begin date.
Though, with most languages, I would say all we can do, as far as talking about it's age, it talk about how far back it would be mutually intelligible. Which if one considers spoken language primary, takes some guesswork.
豊川祥子 said,
April 1, 2025 @ 10:06 pm
> I think, with Mandarin and Arabic, if we are talking about the standard forms, we can date their creation and say they are young languages. Not created ex nihilo like some sign languages, but, still, something we can mark as a begin date.
On what minute of what hour on what date did Standard Mandarin drop then? And where did it come from? In what sense is this situation different from Cantonese's (perhaps former, fading) status as the lingua franca of the Pearl River Delta? How many of the grandparents and great-grandparents of contemporary Cantonese speakers were native speakers of prestige Cantonese?
the comparison with Modern Standard Arabic isn't really sound at all because MSA has no native speakers.
Thomas said,
April 1, 2025 @ 11:38 pm
For the mildly language-educated layman, I usually compare the situation to the Romance languages. Starting from Latin, French seems to be the most degenerate in that it sounds like Latin was chewed to death. But in comparison to the phonologic innovation, the language is trapped in some ancient way of writing. And arguably, French is the Romance language with the highest prestige. Now Mandarin is the French of Chinese. Some other variety can then take the position as most conservative.
Julian said,
April 2, 2025 @ 2:40 am
Okay, it looks like by 'older' we mean 'has changed less since the time of the most recent common ancestor.'
In that sense, chimpanzees are older than human beings because the most recent common ancestor probably looked more like a chimpanzee than a human being.
That terminology is admittedly awkward. It's probably better to say 'X is closer to [='looks more like'] the most recent common ancestor than Y is' rather than 'X is older than Y'