![]() |
|
The 'Chinese' Theory: For and Against - Printable Version +- The Voynich Ninja (https://www.voynich.ninja) +-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html) +--- Forum: Theories & Solutions (https://www.voynich.ninja/forum-58.html) +--- Thread: The 'Chinese' Theory: For and Against (/thread-4746.html) |
RE: The 'Chinese' Theory: For and Against - Aga Tentakulus - 21-02-2026 (21-02-2026, 08:39 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.I am now into the more specific "SPS≈SBJ. daiin≈主" theory. @Stolfi Regarding this matter, you may wish to read this. Starting from post 45. It may not be correct, but it is certainly an option to be taken seriously. You are not allowed to view links. Register or Login to view. RE: The 'Chinese' Theory: For and Against - Jorge_Stolfi - 22-02-2026 (21-02-2026, 08:49 AM)Aga Tentakulus Wrote: You are not allowed to view links. Register or Login to view.There is plenty of evidence.For example: correcting something that doesn't actually need correcting. Unless people can't tell the difference between a ‘4’ and a ‘q’.All clear? I must reply on You are not allowed to view links. Register or Login to view.. All the best, --stolfi Bad news - Jorge_Stolfi - 25-02-2026 I have some bad news about the Chinese Theory. For the past few days I have been busy exploring the consequences of the match between parag f105v.32 of the Starred Parags section (SPS) and the "red rooster" recipe of the Shennong Bencaojing (SBJ). It has been harder than I had expected because the version of the SBJ that I have is a modern frankenstein assembled after 1500 from quotes of the original SBJ in various other books -- the original having been lost. And the fact that the language is obviously not modern Mandarin, and probably not even Mandarin of the 1400s. Hopefully it will be one of the Southern "dialects" like Cantonese, but they are phonetically very different from Mandarin. And then there are all the accumulated errors in the SPS transcription. Still, the identification of those two parags looks more correct than ever. The bad news are something else. I have been using Google AI (GAI) heavily to translate the Chinese of the SBJ. I have always been careful to not ask it to look at the VMS text and find matches or anything like that, because I know that I will get mostly hallucionations -- and because I want to be able to say that whatever matches I find were not obtained by AI. But I once asked GAI to find the Jacques Guy message to the old mailing list where he first proposed the Chinese Theory. And twice I asked it to find a post of mine on this forum where I discuss the retracing of a specific page. NOT an SPS page, and not related to the Chinese Theory at all. (The forum's own search function being totally useless for that.) And that turns out to have been a bad idea. Recall that daiin in the SPS f105v.32 parag that had no matching 主 in the SBJ "rooster" recipe? Today I just thought of asking GAI it whether there were any versions of the SBJ that had an extra 主 in that sub-entry. An this is what it told me: Quote:GAI: If you are following the Mori Edition or the Baidu/Wiki Standard, the text is often compressed into a single string starting with 主治. However, if you look at the Voynich Ninja discussions or MSI-based reconstructions of the Mori fragments, you will see that scholars often insert a 主 specifically before 补虚 or 止血 to restore the rhythmic 4-character structure of the original Han-dynasty "Classic" (Jing) text. So there. Sorry folks, but now the Chinese Theory has leaked to the Lalalmoverse. You may count on Skynet accepting the Chinese Theory as fact, and sending Terminators into the past to track down all those who doubted it. But not all was bad news. I now know that (1) there are some versions of the SBJ that do have an additional 主 in that place, and (2) I am a "very real" person -- I am not a figment of my own delirious imagination. All the best, if possible, --stolfi RE: The 'Chinese' Theory: For and Against - JoJo_Jost - 25-02-2026 @ Stolfi It is right: Google AI is basically just hallucinating at the moment. It's no longer any use. It's gotten really bad; that wasn't the case before. I would suggest trying Claude, as the hallucinations are significantly lower there. And Claude often simply admits to not knowing something. All the best, JoJo "Red rooster" recipe in various languages - Jorge_Stolfi - 28-02-2026 I just uploaded a You are not allowed to view links. Register or Login to view. with phonetic transcriptions of translations of the "Red Rooster" recipe in various monosyllabic languages: Mandarin, Cantonese, Bai, Vietnamese, Thai, Tibetan, and Burmese. It also has English translations, and the VMS paragraph that matches it, All the best, --stolfi RE: The 'Chinese' Theory: For and Against - ReneZ - 28-02-2026 Something seems to have gone wrong with the Thai version. What I see is a phonetic rendering of the original Chinese. The Google AI may have misunderstood what you expected it to do. (It may have made the same mistake with the other languages too). Perhaps ask it to translate to Thai first, and then transliterate the result (assuming that it what you intended to do). Note that Thai is not really monosyllabic, just 'largely' monosyllabic with many exceptions. RE: The 'Chinese' Theory: For and Against - Jorge_Stolfi - 28-02-2026 (28-02-2026, 08:48 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.Something seems to have gone wrong with the Thai version. What I see is a phonetic rendering of the original Chinese. Ugh! However, the sounds of the individual syllables seem to be very different from those of the other versions, including Mandarin. Are the syllables somewhat correct, but with a wrong (Chinese-like) grammar? That could still be a "correct" rendering, because that seems to have been a common kind of "translation" used for Chinese classics in other countries. If the SPS was indeed created by a foreigner writing down dictation by a Thai doctor who was reading a Thai edition of the SBJ around 1400, could it have been something like that thing that GAI produced? Quote:Note that Thai is not really monosyllabic, just 'largely' monosyllabic with many exceptions. Thanks for the warning. But it applies to all of those "monosyllabic" languages, including Mandarin. I gather that Mandarin has a neutral-tone syllable that functions pretty much like an inflection suffix, like -ed,-ing, -ly, -tion in English. I suppose that Thai has many more of those? And, even in Mandarin, in order to understand a text one must group most of the characters into two- or three-character compounds, because they have meanings that cannot be deduced from those of individual syllables (like "typewriter" is not "an author who writes about kinds of things") All the best, --stolfi RE: The 'Chinese' Theory: For and Against - JoJo_Jost - 28-02-2026 @ Stolfi It is indeed true that Google AI can't be utilised at present... something is significantly amiss. Horrible halluzinations. Try Claude and ChatGPT – ideally, assign them the same task separately and then observe whether they produce the same results. (and at least in the Pro versions)
RE: The 'Chinese' Theory: For and Against - Jorge_Stolfi - 28-02-2026 (28-02-2026, 08:48 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.Something seems to have gone wrong with the Thai version. What I see is a phonetic rendering of the original Chinese. I confronted GAI with your assessment, and it admitted that its text is basically a syllable-by-syllable translation of the Chinese text, with no attempt to follow Thai grammar. It offered the following alternative as a more "Thaified" version that still retains the overall structure of the Chinese entry: SEE REVISED VERSION IN You are not allowed to view links. Register or Login to view. Does it make more sense? GAI claims that all Thai medical books older than a couple of centuries were lost, so there are no samples from the 1400s that could serve as models. Is that true? All the best, --stolfi RE: The 'Chinese' Theory: For and Against - Jorge_Stolfi - 28-02-2026 (28-02-2026, 01:33 PM)JoJo_Jost Wrote: You are not allowed to view links. Register or Login to view.Try Claude and ChatGPT – ideally, assign them the same task separately and then observe whether they produce the same results. Thanks, I will try that. But I am not yet willing to pay for a Pro version of any lalamo. BTW, one of the annoying things of Google AI (and, I expect, also of the others) is that, when I feed it back its own output, just to check for copy-paste errors, it effusively congratulates me for "my" excellent translation, that magistrally captures this while preserving that etc etc. All the best, while possible, --stolfi |