Abstand und ausbau, part 2
« previous post | next post »
The first part of this debate, "Abstand und ausbau" (10/28/25), was so spirited and prolonged, and has recently moved on to significant new ground, that I've decided to launch this part 2.
Before commenting here, please go back and review what was said in the previous o.p. and the subsequent comments thereto, some of which are quite substantial. Here I copy one of the recent observations in the first thread that has not yet been adequately responded to there:
The genetic unity of Sinitic is not only unproven, but it's being challenged increasingly by both paleogenetics studies, as well as paleoanthropological studies, including by mainland Chinese researchers affiliated with mainstream state institutions.
The "Western Xia, Eastern Yi" hypothesis, which says the civilization later known as Han came to be through merger of two large civilizations (one coastal and the other inland), each with their own (unrelated and starkly different) language, proto-writing, spiritual believes (shamanism vs. ancestor worship), economic production (rice vs. millet) and material culture, was proposed decades ago, remains somewhat fringe to this day, but is slowly picking up support.
(wgj)
[The commenter appends two videos in Chinese, here and here, the second of which is not currently working.]
One thing is certain: whatever Hànyǔ 漢語 ("Sinitic") is, it will take decades to figure out its synchronic and diachronic dimensions.
Selected readings
- The works of Søren Christian Egerod (1923-1995), particularly his magisterial "Chinese languages" and other articles in Encyclopædia Britannica. I'm surprised and gratified that Egerod's superlative, prescient scholarship is still being republished in current editions of Encyclopædia Britannica. I read his articles in an earlier, printed edition. I am honored to have met Egerod in person in Taiwan four decades ago and to have heard him pronounce his name in Danish, a mind-boggling experience.
- The works of Jerry Norman (1936-2012), especially Chinese. Cambridge language surveys, (Cambridge University Press, 1988). Norman privately told me that the Sinitic group had more than three hundred different languages.
See also the bibliography at the conclusion of the first part of this series.
Jerry Packard said,
November 11, 2025 @ 10:11 am
“The genetic unity of Sinitic is not only unproven, but it's being challenged increasingly …”
In many ways the genetic unity of Indo-European or PIE are also unproven and subject to more detailed future data analysis.
wgj said,
November 11, 2025 @ 11:43 am
Here's the link to the second video again:
冯时 文字起源与夷夏东西 20200712
https://youtu.be/bKFHZ6biVCw
It's a recording of an online seminar (during Covid), in which a presentation on the origin of Sinitic writing the "Western Xia, Eastern Yi" hypothesis was given by Prof. Feng Shi with (supporting) comments by other experts. Like I said, this hypothesis is still considered fringe, but Prof. Feng's background and affiliation are as mainstream as it gets, graduate of Peking University's Department of History (which used to include archeology before it was split off) and working at Institute of Archeology of the Chinese Academy of Social Sciences, the two top authorities in Chinese archeology.
wgj said,
November 11, 2025 @ 12:14 pm
@Jerry: Unlike Sinitic which has been geographically limited until modern times, Indo-European has been so widely spread that it's seems statistically almost certain that at least some IE language would have been a non-genetic cultural adaptation – meaning some group not directly descended from PIE speakers must have taken on IE as a replacement of their previously non-IE language, the way African-Americans have taken on English.
What I've been wondering for a while now: Weren't the Franks originally Germanic? When and how did they drop German and take to Latin instead?
Victor Mair said,
November 11, 2025 @ 1:41 pm
@wgj
You've given us the first video again, and it always worked.
Here's the URL of the second video, and it still doesn't work:
https://youtu.bexZv_FgqgqrA
Philip Taylor said,
November 11, 2025 @ 3:08 pm
For the latter, please try https://www.bilibili.com/video/BV1qf4y1m7mL/ — caveat lector : I am not able to guarantee that this is indeed the same video.
Philip Taylor said,
November 11, 2025 @ 3:09 pm
P.S. The original YouTube URL is lacking a slash, but even once interpolated the content is no longer available — https://youtu.be/xZv_FgqgqrA
VMartin said,
November 11, 2025 @ 4:44 pm
@wgj.
Nikolay Marr, the former head of Oriental studies during Stalin claimed that the concept of IE languages is only a bourgeois propaganda. One of his arguments was that French doesn't have declension. Another one was, that IE languages have different words for basic words like good – bonum, хороший, gut, dobrý… Stalin later criticized Marr's concept of the class origin of languages, and modern linguistics has rejected his teaching entirely. Yet it is said he spoke some 20 languages, and his work is now available in Russian in several volumes. Wrong or not – for an amateur like me it is an interesting reading. And it supports your claim, that IE was a language of the ruling class of conquerors implemented on pre-existing languages in Europe.
wgj said,
November 12, 2025 @ 12:31 am
Here's the other link:
古DNA解析-东亚人群8000年来的迁徙和混合历史
(Paleogenetics: The history of migration and mixture of East Asian populations in the past 8000 years)
https://youtu.be/xZv_FgqgqrA
Yves Rehbein said,
November 12, 2025 @ 3:34 am
@ wgj, I would say Frankish is too fringe, because it is hardly attested. First you would need to talk about Gallo-Romance on the one hand and, on the other hand, the possibility of an already Proto-Germanic dialect chain.
As for Sinitic, Matisoff's review of the Cambridge Handbook of Areal Linguistics points out positively that Hilary Chappell's contribution distinguishes five linguistic areas. The chapter does not say much about ancient history, though.
My impression is that few would dare to challenge the Anyang inscriptions. I read that "… the Three Dynasties resided in between the Yellow and Luo Rivers, with Song prominent as the central peak, with the four other peaks each off in their own directions, and four rivers all on the east of the Mountain" (The Han shu 漢書 cited after Qi Wenzin 2017 [translated by Susan Blader] in FS Sarah Allan vol. 3) and the author precedes this stating that Mt. Song "… is also the core area from which Xía, Shang, and Zhou civilizations emerged and spread." (Qi Wenzin 2017). Another fourfold division — including Yi — can be gleaned from Victor Fong 2025 in JAOS.
We can be fairly sure that there were bamboo slips, because the 冊 "book" character exists, yet there is no record of them, so any argument about uniqueness is moot if you have to assume that slips existed elsewhere. There may be a relation to 辫 "queue, braid", exopassive of 編, phonetic 冊, if queue in Latin coda (cauda) is also related to codex "book" (caudex).
wgj said,
November 12, 2025 @ 5:55 am
Anyang isn't really relevant here because it's much too young. We're talking about the time frame between 10K and 4K years before present (the Jiahu site in Henan is 9+K BP and has already produced 100+ proto-writing symbols). Within the Xia-Yi hypothesis, there are further speculation that Chiyou might be the old god of Yi – in fact the Yi people in SW China today (which may or may not be descendants of the same Yi as in Xia-Yi) does worship Chiyou as an important god of theirs, dedicating festivals to him still. Or else there are speculation that Yu (AKA Dayu) didn't so much tame the food, but actually conquered an aquatic people, known today from the Liangzhu site outside Hangzhou, which is ca. 5K BP and also the oldest known rice cultivators (and there are evidences of a very sudden decline of the culture, possible by natural desaster or by war). The absolutely out-there theory says that Yu in fact manufactured the flood, as a WMD against the Liangzhu people, by damming the Yangtze upstream and creating a reservoir, before breaking the dam and flushing the water down. There are enough tentative evidences to prevent those outlandish speculations from being rejected out of hand, but to actually demonstrate them to a scientific standard is an entirely different matter.
Chris Button said,
November 12, 2025 @ 6:55 am
Pulleyblank suggested that 夏 and 華 might be variants. Phonologically it's not unreasonable.
wgj said,
November 12, 2025 @ 7:23 am
Also relevant to our discussion on possible connections between Sinitic and Indo-European is a particular interpretation of the story of three legendary figures, Flame Emperor, Yellow Emperor, and Chiyou.
Common (non-fringe) legend has it that the two emperors struck a deal and merge their peoples, then combined forces to defeat Chiyou. Now, the fringe theory says that Flame Emperor was the leader of a people from the West, descended from Ancient Iranians – and therefore PIE speakers! They worshipped fire, as did their relatives who stayed in Persia and created Zoroastrianism, which is why their leader was called Flame Emperor.
Yellow Emperor was presumably native to the Northern Chinese Loess Plateau, his name reflecting the color of the loess. Chiyou, whose name is interpreted by some scholars as "Destructive Loon" (obviously not an endonym, but a Trumpian insult issued by his conqueror), was the leader of the Yi people – a coastal folk whose culture was even more alien to the Yellow Emperor than the distant travellers from across the continent (!), and therefore must be destroyed rather than reconciled.
If we were to accept this theory prima facie, then the Proto-Sinitic speakers (Yellow Emperor) absorbed the PIE speakers (Flame Emperor) first, and (semi-)peacefully, then the Ancient-Yi speakers later, and violently. Which would mean that elements from PIE might indeed be better preserved in Sinitic than those from Ancient-Yi.
I personally find this whole thing too convenient by half. But it's certainly interesting, and definitely entertaining.
wgj said,
November 12, 2025 @ 7:50 am
I support the view that before we ask what the origin of Xia is, we need to ask whether the people/dynasty was named after the season of summer (or something related to it, like the cicada, which some scholars have interpreted the oracle script for xia to resemble), or vice versa. It's a similar question to that regarding Shang: Are they called so because they were known as merchants, or was the profession of merchants named after them?
Many Chinese historians and paleolinguists believe that ancient Chinese didn't divide the year into four seasons, but only two – spring and autumn (therefore the title of the famous book Spring and Autumn basically meant "all year long", "year after year", or "chronicles"). Summer and winter were carved out as new seasons later, which meant that the season xia was likely named after the people/dynasty Xia. If that's true, then xia might very well be cognate to hua (meaning blume).
Speaking of Shang, there are some followers of the Xia-Yi hypothesis who believe that the Shang tribe, originating around today's Shandong, was descendants of Yi (after it had been subsumed into Xia). That would mean the overthrow of Xia by Shang constituted a retribution for the conquest in the other direction hundreds or even thousands years prior. Talk about revenge best served cold!
wgj said,
November 12, 2025 @ 8:54 am
There are four neolithic city-states generally recognized in Chinese archaeology:
– Liangzhu outside Hangzhou (5300-4300 BP)
– Taosi in Shanxi (4300-3900 BP)
– Shimao in Shaanxi (4300-3800 BP)
– Erlitou outside Luoyang (3900-3300 BP)
Among those four, Liangzhu clearly stands out in every aspect – the oldest, the largest in size (by far), the only one south of the Qin-Hua climate divide, the only aquatic culture (and rice farmer), etc. Followers of the Xia-Yi hypothesis mostly believe it was Yi, whereas the other three were Xia. The downfall of Liangzhu, before the others even reached civilizational status, is therefore interpreted as an advanced culture (Yi) being conquered by a comparatively primitive but militaristic one (Xia). You can see why Chinese people, with their history of being plagued by waves after waves of people from the steps, find this narrative alluring. :D
Another point of interest is Erlitou, which is usually classified as belonging to the Longshan culture, based on material culture. For those who believe that something akin to a dynasty named (or later referred to as) Xia existed, Longshan (in archaeology) is roughly equal to Xia (in history). The Longshan site itself (4500-4000 BP, featuring no ancient city) is outside Jinan in Shandong, and considered early to middle period of Longshan culture, whereas Erlitou represents late Longshan culture.
However, the material culture from the Longshan site has a multitude of features not shared with other cultures across the Northern Chinese Plain (even though the similarities are greater than the differences), in particular the stunning black egg-shell-thin pottery, which has led some to speculate that the Longshan site itself was indeed a Xia-Yi mixed culture.
Erlitou is generally interpreted as the very (bitter) end of Xia (at least for those who accept Xia's existence). The palace complex was destroyed in a fire, then instead of repairing and rebuilding it, a new complex (now referred to as the Shang city) was (almost) immediately built just five kilometers away. Why such a waste of resources, not to mention political continuity and legitimacy, if not to send a strong message: A clean break – out with Xia, in comes Shang! And if you believe the Shang were Yi (see above), then the (Yi) empire has struck back …
wgj said,
November 12, 2025 @ 9:20 am
@Yves: Sorry, I wasn't clear before. My question about the Franks and their language switch wasn't in reference to Frankish, but French! Because France was built by the Franks. Were they already speaking proto-French (i. e. a Latin language) when they moved to Gaul? Or were they still speaking Frankish (i. e. a Germanic language) and only changed later due to population dilution by the natives?
Victor Mair said,
November 12, 2025 @ 9:51 am
The traditional designation "Huáxià 華夏" should be taken into consideration.
For an in-depth discussion of the alleged historicity of Xià 夏, see "Was There a Xià Dynasty?", by,Victor H. Mair, with contributions by E. Bruce Brooks, Sino-Platonic Papers, 238 (May, 2013), 1-39, with additional notes by numerous scholars in Sinology and related fields. (free, easily accessible pdf)
wgj said,
November 12, 2025 @ 11:03 am
Thank you for that reference, but I think it's fair to say that academic research on Xia has moved on quite a bit since 2013. In particular, I consider the opening of the Erlitou Site Museum of the Xia Capital in 2019 a monumental milestone in the history of this research. Because for the first time, an official state institutions is making the claim for the existence of Xia – right in its name, no less. And they were hyper-aware of the lack of scientific consensus on this point, which made them extreme diligent in arguing for their case. This museum is the only one I've visited – not just in China, but anywhere in the world – where every explaination, every text and graphic comes with citation (a scientific paper or book). It's a culmination of decades of work by generations of experts – and it's still ongoing. When I visited again last summer, I was surprised to see that in the archaeological site (the museum is built next to it), one already backfilled ditch was dug up again, and a completely new ditch was being excavated.
Victor Mair said,
November 12, 2025 @ 2:17 pm
@wgj
Are you saying that scholarship published more than a decade ago that deals with historical materials dating back thousands of years is no longer to be trusted?
Chris Button said,
November 12, 2025 @ 5:07 pm
There were definitely loanwords. The idea that the language families sprung from a common source is what's debatable. The structural commonalities identified by Pulleyblank are perhaps more reflective of how language works in general than of a common source.
Incidentally, Pulleyblank thought PIE "e" was schwa and "o" was /a/. Conversely, Kortlandt and Matasović invert the relationship to treat "o" as schwa and "e" as /a/. I think they are probably correct to invert it.
I think you might be referring here to an old misanalysis of the oracle-bone character signifying 秋 as 夏.
Jonathan Smith said,
November 12, 2025 @ 5:57 pm
* the question of whether a language is or isn't "IE" or "Sinitic" or whatever is to be addressed (or not) with language data and has nothing to do with population structure
* there are lots of vague ideas about the cultural components that contributed to whatever "Chinese" civilization is: Huaxia >> "Chinese" is kinda one, Western Xia + Eastern Yi >> "Chinese" is kinda one, (components of) Northern Di + Eastern Yi + Southern Min + Western Rong is kinda one… needless to say none of these are remotely edgy or sophisticated. Thankfully there is archaeology.
* there are lots of vague ideas about the nature of "Proto-Sinitic" should one choose to accept it: it's a TB sister, it's a TB branch, it's a mixed language, it's a creole, etc., etc. Thankfully there is linguistics, in particular the possibility of studying and comparing modern Chinese-ish languages independently of any of these hypotheses.
* the early ancestors of the modern glyph "夏", like other seasonal graphs, would have written names of seasonal periods or phenomena but also other (conceptually related!) words. The OBI bug character is an ancestor of neither modern "夏“ nor modern "秋". Probably none of this is directly relevant to the notion of a "Xia" dynasty…
Chris Button said,
November 12, 2025 @ 6:18 pm
@ Jonathan Smith
The 'bug" character was phonetic in the character used to represent the word now represented by 秋 "autumn". On its own, it did not mean "autumn" but some kind of pest that needed to be pacified.
Jonathan Smith said,
November 12, 2025 @ 6:30 pm
@ Chris Button
No but by all means write it up / publish it…
Chris Button said,
November 12, 2025 @ 6:51 pm
Respectfully, you could just read the already published literature on the topic?
I'm also not clear what you are taking issue with since your statement that the "bug" is not an ancestor of 秋 is entirely correct. And 夏, as you note, has nothing to do with it.
It is, after all, the pest (not autumn!) that is 寧-ed in the oracle bones.
Jerry Packard said,
November 13, 2025 @ 8:45 am
“As for Sinitic, Matisoff's review of the Cambridge Handbook of Areal Linguistics points out positively that Hilary Chappell's contribution distinguishes five linguistic areas. The chapter does not say much about ancient history, though.”
Chappell (Ed., 2001) _Sinitic Grammar: Synchronic and Diachronic Perspectives_, offers 13 articles, mostly involving grammatical markers and inflections, many of which describe the evolution of such markers over time, including evidence from the modern dialects.
Chris Button said,
November 13, 2025 @ 8:45 pm
This seems to be borne out in the oracle-bone inscriptions. Two seasonal words are attested:
– 春, which is now used for "spring"
– 秋, which is now used for "autumn" (as discussed above)
Jonathan Smith said,
November 13, 2025 @ 11:08 pm
Tragically the (relatively speaking!!) coherent literature on this topic is by me (and don't hold your breath for better), so I can say with confidence that "BUG" (although careful because similar but different glyphs are mixed together in casual discussions!) writes a seasonal term in the OBI and that asserting that the term in question must mean 'summer' or 'autumn' or be etymologically related to seasonal terms of e.g. modern Mandarin or be graphically related to modern "夏" or "秋" is… is… (…numerous edits for manners…) not right. FWIW the latter two glyphs have their own fascinating and totally separate histories. And there are other seasonal terms in the OBI… you must look at the original materials with fresh eyes and select secondary literature extremely um selectively however.
Chris Button said,
November 14, 2025 @ 8:07 am
@ Jonathan Smith
I'm not sure why you are still talking about 夏. By the way, the character 冬 is attested in the oracle-bone inscriptions, but it is not used for a season there.
As for 秋, start with work by titans such as Ken Takashima and Qiu Xigui.
The 龜 component in old variants for 秋 like 䆋/ is a graphic corruption of the "pest" (bug) that is 寧-ed ("pacified") in the inscriptions.
"Pest" as phonetic with 禾 or 火 occurs with the meaning of a season in the oracle-bone inscriptions. It is a counterpart to another season 春. Nowadays, 秋 and 春 are specifically used for "autumn" and "spring". The "pest" graph also occurs as straight-up loangraph for 秋, where context makes it clear that it is referring to a season rather than a pest.
Sean said,
November 14, 2025 @ 9:40 pm
wgj: the Franks wondered this themselves and by around the year 1000 had made up a story that obviously their ancestors first made the Gallo-Romans teach them Latin, then killed them all. Fun-ruining historians noticed that since Franks got to wear swords and speak in the assembly, but Gallo-Romans had to pay taxes and be humble (and since they had important skills like reading and writing and serving mass) over a few generations most families in northern Gaul discovered that they were really Franks. Its very common that a small military elite like that loses its native language, Nicholas Ostler has case studies.
KIRINPUTRA said,
November 15, 2025 @ 12:08 am
"Genetic" means something else in linguistics.
The linguistics term is counterintuitive. It's partly our fault that non-linguists miscontrue it. It's not a well-chosen metaphor and even linguists seem confounded by it sometimes.
Jonathan Smith said,
November 18, 2025 @ 6:09 pm
For anyone who cares (pipe down crickets), re: paleographical problems pertaining to early seasonal cycles and formative calendrical astronomy check out e.g. my "The old astronomical significance of the glyph 歲" or "Dong 東 'east' and the Chinese 'Indian Circle'" — with the first few pages of both devoted to pointing out some of the "field's" (cough, hack) astonishingly ill-founded conventional wisdoms.
@Chris Button I guess you are referring above to the specific conventional wisdom that (OBI) PLANT-LIKE-THING-ABOVE-MOUTH represents a word 'spring' and is in some way or other related to modern "春" and that BUG-WITH-PROMINENT-ANTENNAE represents a word 'autumn' (earlier idea: 'summer') and is in some way or other related to modern "秋" (earlier idea: "夏"). For the last time: these are hugely typical but prima facie silly attempts to force inscriptional evidence to match up with "Chinese Literature & Culture!!" / modern character forms / words of modern Mandarin. What can I say, I am as sorry as anyone that papers like those named above represent the state-of-the-art. Update your understanding or not, and test your claims against anonymous reviewers or not, as you will. If "not" — as has been the case for going on two decades (!) — by all means continue lamenting on the Internet that No One Gets Anything except you.
Chris Button said,
November 18, 2025 @ 7:54 pm
@ Jonathan Smith
I'm sorry you feel that way. I can but defer to the expert palaeographers.
As for the 春 season, the connection noted by experts like Qiu and Takashima is rather with 屯 than the character you are describing.
Yves Rehbein said,
December 13, 2025 @ 5:51 pm
@ wgj, Frey and Salmons (2012, in FS Mallory) briefly refer to Roberge (2010:418-19, in Handbook of Language Contact, see 2nd ed. 2020:332), though the paragraph is not precise enough to answer your question about the Franks. More importantly, the era that leads up to this point is somewhat obscure:
(Roberge 2020:330).
It follows that the comments on Frankish are necessarily superficial. I am just disappointed that the cited range over two pages promissed more. Unfortunately, I do not have the first edition to confirm that. The key-word seems to be "vertical" (compare Frey and Salmons 2012) in the sense of socio-linguistics.
(Roberge 2020:332)
I was rather thinking of Burgundian, since the recent issue of NOWELE mentions it in the title of Haubrichs's article: "Burgundische Personennamen im Kontakt mit Fränkisch und Galloromanisch als Elemente der Rekonstruktion einer ostgermanischen Trümmersprache". The fact that an English speaker can read the title in German probably is a case in point (cf. Average European Dialect continuum). Note the replacement of Middle English mid ("with") with with. However, the "ostgermanisch" label clashes with Kim's "On the Phylogentic Status of East Germanic" (2023, in FS Ringe), which concludes: "First, the label 'East Germanic' has no cladistic value …" (Kim 2023:38). At this level of debate it seems uninviting to base a typological case study on that, although the need for interdisciplinary discourse is arguable.
Yves Rehbein said,
December 13, 2025 @ 6:14 pm
Correction: *[Standard] Average European, SAE for short. I have not read anything about it, just saying.