I had hoped analogy would prove useful, and that I would get a direct answer to a direct question, but it seems not.
Let's look instead at this claim:
(20-06-2026, 07:53 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.The situation seems to be the same as that of Serbian and Croatian, only that in reverse. AFAIK, linguists agree that the spoken forms are essentially the same Serbo-Croatian language, which has a Croatian spelling using Latin letters and a Serbian spelling using Cyrillic. But I gather that some people, especially in Croatia, insist that they are completely different languages -- again, for political motives.
To my understanding the example is correct. But it is an entirely different situation from Chinese.
The reason a person in Guangzhou can read a newspaper from Beijing and vice versa is that there is mostly one written language in China, based on Mandarin, and literacy is synonymous with knowing Mandarin. I will call this Standard Chinese, but it has a few other names. I do not know why I am the first person to tell you this, but it is a readily available fact.
You are not allowed to view links.
Register or
Login to view.:
Quote:It is commonly claimed that there is only one "Chinese" language, and that all of the variants of that language are dialects of it. This conception of there being only one "Chinese" language plays havoc with efforts to classify the countless varieties of Sinitic speech forms into meaningful groups, branches, languages, and dialects, as is normal for other large families or groups of languages.
The old canard that "when the dialects are written down they are the same" is simply untrue, since what gets written down are not the regional variants but standard Mandarin (and in earlier times Classical Chinese, a dead language for at least two thousand years). If one, as a tour de force, does contrive to write unadulterated Cantonese or Taiwanese, for example, they will be as hard for a reader of Mandarin to understand as spoken Cantonese or Taiwanese is for a speaker of Mandarin to understand.
I say mostly, but there is "You are not allowed to view links.
Register or
Login to view.", and its very existence should cast doubt on the idea that the differences are just dialectical. From the Wikipedia page:
Quote:While the Mandarin form can to some extent be read and spoken word for word in other Chinese varieties, its intelligibility to non-Mandarin speakers is poor to incomprehensible because of differences in idioms, grammar and usage.
It goes onto say:
Quote:The standardization and adoption of written Mandarin preempted the development and standardization of vernaculars based on other varieties of Chinese. No matter which dialect one spoke, they still wrote in standardized Mandarin for everyday writing. However, Cantonese is unique amongst the non-Mandarin varieties in having a widely used written form. Cantonese-speaking Hong Kong used to be a British colony isolated from mainland China before 1997, so most HK citizens do not speak Mandarin. Written Cantonese has developed as a means of informal communication. Still, Cantonese speakers must use standard written Chinese, or even literary Chinese, in most formal written communications, since written Cantonese may be unintelligible to speakers of other varieties of Chinese.
(I think this sells a few of the other written varieties a little short, but depending on the cut-off for "widely used" and how much weight you put on the situation in Taiwan, it is not glaringly
wrong either.)
You are not allowed to view links.
Register or
Login to view.:
Quote:Much of the time "written Cantonese" is "Mandarin written by a Cantonese speaker" and so can only be distinguished from native Mandarin by relatively subtle cues. However, if what is written is truly Cantonese, it is easily distinguished from Mandarin if it is of any length because some common words are not cognate and are therefore written with different characters.
You might notice a parallel with your description of medieval Latin, though I would not encourage you to describe living people using a lingua franca as speaking "brokenly". Again, the main form of written communication in China is
Mandarin, barring some critical caveats.
It is easily verified that this is an educational choice, that the language of instruction in Chinese classrooms is not the vernacular, but rather Standard Mandarin. The situation You are not allowed to view links.
Register or
Login to view., and this is very much by design. As shown, the mutually unintelligibility is not merely due to sound change, but due to far-reaching differences between these languages, so this is not about children just speaking with Beijing's pronunciation, but an entirely different language being learned.
I have seen parallel claims made about Classical Chinese, where it is even more preposterous. Bill Bryson, in his quite enjoyable
English and How it Got That Way, claimed that people in China can read their classics "as if they were a shopping list". In Guangzhou, it is sociologically accurate to say Cantonese speakers can largely read Beijing newspapers; I've shown this is a description of widespread bilingualism, not something arising from the linguistics of Cantonese, but it is largely accurate. Classical Chinese is on the national tests, not because it is transparent to students but because they are mandated to learn it in addition to Mandarin. I don't want to understate the transference benefits the logographic script gives when students undertake this, and I am personally familiar with it from my own language studies, but I am not alone in thinking it is commonly
You are not allowed to view links. Register or Login to view. in the West. From what I can tell, the most intensive parts of that education are after mandatory education ends in 9th grade, so it's not like every lettered person can just casually pick up 2500 year old texts and read them like a shopping list. At any rate, insofar as there is some truth to the claim, it is a description of what languages Chinese people formally learn in school rather than a linguistic description of written Chinese.
And just, gut check here! The quotes by linguists and the descriptions of the educational landscape should be what convinces anyone reading this that written Cantonese, written Mandarin/Standard, and written Classical Chinese are not mutually intelligible, but a quick look at the time depths here should cast doubt on the common claim to the contrary. Classical Chinese was spoken* 2500 years ago. Mandarin and Cantonese diverged 1500 years ago. The implicit claim that there has been little enough grammatical and semantic change so as to render all Sinitic languages mutually intelligible in writing should raise more suspicion, I feel. I realize that doesn't quite get us to proving it's false, but I don't think people fully appreciate what is being claimed here.
In fact, it would be as if every literate Romance speaker could fluently read Classical Latin, which someone might exactingly separate from Vulgar Latin, and any other Romance language that has evolved since the time of Cicero. A major hurdle to this in Europe is that national languages have their own spellings, but this is easily fixed. We can take the Latin spelling and just leave each word unchanged. I understand this is a perverse thing to do given that the logic of an alphabet is to represent phonemes, but I admit I did not think that would be quite as big an ask of my audience as it turned out to be. I also did not think, given the knowledge of Latin on this forum, that someone would insist there is an inherent pronunciation of written Latin given that historically there were many, many ways to realize a Latin text orally, and even today there are two major ones still in use. (Also, it is deeply amusing to me, a native English speaker, to consider the Latin script as having an inherent phonology, even if that's slightly different from what you are saying.) If you fix the spelling of
words instead of
phonemes in the time of Cicero, and then write French with that orthography, what you get is, "Illi non me placet passum de totus." As ahistorical and frankly illogical as this would have been, I also don't think it's
that hard to understand that <pas> would be <passum> if all those historical choices you mention were made differently and instead French writers had decided to protect Latin orthographical words when transitioning to vernacular writing.
The point is that this is
not a perverse thing to do with a logography, so much so that explaining it in writing looks a little silly: the way you write 雄 in Classical Chinese is <雄>, the way you write it Mandarin is <雄>, the way you write it in Standard is <雄>, and the way you write it in Cantonese is <雄>. What this does not tell us is pronunciation
or usage. Usage of 雄 happens to illustrate one of the ways you have to be careful treating Standard as strictly identical to Mandarin; the precise usage in "rooster" is borrowed into Standard in some contexts directly from Classical Chinese, but it is not correct in colloquial Mandarin. Not only is the colloquial compound for "rooster" in Cantonese different from Mandarin, it also doesn't owe to the Classical construction. This is directly analogous to the observation that French <pas> is <passum> but for lexical, grammatical, sound, and orthographical changes. There's a way in which they are the same word, and my "classical French orthography" leverages this fact in exact parallel to Chinese, but multiple people quickly saw there are important ways they are not the same word. For most of your audience, the parallel with the "Mandarin" and "Cantonese" on You are not allowed to view links.
Register or
Login to view. is opaque, but what we might call the passum-pas problem is frequent in this kind of reading**. I've also repeatedly mentioned the homophone problem, which has a little bit of an analogy in Modern French, but was so bad it drove a lot of the changes between Classical and Middle Chinese.
The reason I've repeatedly said the readings you've proposed look fine to me is that this is a perfectly normal thing to do when reading Classical Chinese; unlike with Latin, the reconstructed pronunciation has little currency outside of specialist research into historical Chinese sound change. At this point, most people use Mandarin readings, and as mentioned a lot of people have exposure to doing these kinds of readings in a classroom. This is not the situation you imagine your author in, where he never had the logographic cues and he had no exposure to the Classical language. In fact, he is outright attempting evade those two crucial things. This is why your hypothetical looks so strange to me. I can imagine a very stubborn or misguided man taking down the readings for hundreds of pages, though that cracking sound you hear is my credulity coming under immense strain. It is an outright misunderstanding of the Sinitic languages to suppose he had much hope of ever recovering meaning from from such an exercise, and I think this would have quickly become obvious to him. That Google's LLM is telling you otherwise would impeach the credibility of Google AI if I thought
things could be credible.
Briefly about translation, the exact path of language change, the proliferation of compound forms after the Classical period, means your statistical arguments are all undercut if you try and imagine the Dictator was accommodating him that way. The SBJ translated into Cantonese, a thing that would have been extremely out of step with practice in 1400 and still isn't really done, would be a very different text than your reference text with different statistical properties. I've been clear that I'm not very convinced by the identification done this way:
(16-06-2026, 02:25 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.The evidence I have says that that what the Dictator did was either (A) or very close to it. The spacings between the cribs are usually very precise, at ~5 EVA letters for each hanzi of the ZHB text.
but you seem to think I should find it convincing. Were I convinced, I would think we could rule out any significant quantity of translation, the (B) option in this scenario. If you think there is a lot of room for interpretation and positing a very different base text than the one you claimed to have identified, I think that's very interesting and you might want to review some of the comments about the robustness of this method by me and other people with that in mind. To be clear, this is not me nitpicking some small deviations, which are quite expected in a premodern text, nor me misunderstanding there are a multiplicity of potential
readings, regardless of
language, that should be consistently identifiable with the same crib in the correct source text. This is me knowing that Classical Chinese is not anywhere close to a one-for-one reflex with languages after ~500 CE, and that your argument accounts for that by erroneously assuming the problem away.
In conclusion, I think you are wrong about how written Chinese works and how it relates to the spoken languages. It is a commonly printed error in the West that Chinese speakers can all understand each other's written languages, a fact that is well-known among people informed about the actual linguistic situation in China. The grain of truth is that most people can read each other's writing, but only because literacy nowadays is almost entirely in Mandarin. It is even easier to exaggerate Classical literacy in the 21st Century, but it is not insubstantial, and that too is a fact about
education rather language. It is controversial to say that Classical Chinese was easily understandable to the illiterate populace in the time of Confucius, and by the 1400s the days that one could understand it aurally based on the vernacular had long passed regardless of ones views on Old Chinese reconstruction. While that does not absolutely rule out "The Author" having taken dictation in Classical Chinese with a vernacular reading, it does make the COT as you've presented it
far less plausible than it might seem based on those common misapprehensions. While it solves the problem of learning the characters, it does not solve the problem of actually reading the Shennong Bencao Jing, which I would take to be at least as important. I understand your position, but it is based on a mistake that raises a lot of questions about the usefulness of the VMS if you turn out to be correct about the text anyway.
*The distinction where Old Chinese was spoken and Classical Chinese was written is usually motivated by a You are not allowed to view links.
Register or
Login to view. about whether or not Classical Chinese was substantially the You are not allowed to view links.
Register or
Login to view., rather than a strict definition about medium. I lean towards thinking this may be correct, thought not only do I think the distinction is unmotivated in this conversation, when speaking of the
written forms I do mean they are cognate between
written Classical Chinese and later forms, even if I'm playing fast and loose about whether or not the written form is strictly the direct ancestor. That said, I have definitely seen sources say "Classical Chinese was the written language", and as a description of usage, that's correct, but hopefully its coming into focus that's a description of Classical Chinese having been the language of writing for 2400 years, not a well-justified description of written mutual intelligibility between well-formed Classical writing and Mandarin Chinese.
**Google AI may say this problem does not exist, but until Google invents an LLM that can accurately explain how it interpreted a prompt and generated its output, there is simply no way to evaluate its "claims", and so the less its "conclusions" are introduced into human conversations, the better. It was a sin to create a machine in the likeness of human mind.
