• Options
• Login
• Desktop Style

The 'Chinese' Theory: For and Against

• Index

  • The Voynich Ninja
  • Voynich Research
  • Theories & Solutions
  • The 'Chinese' Theory: For and Against
• The 'Chinese' Theory: For and Against

• RE: The 'Chinese' Theory: For and Against

  Jorge_Stolfi > Yesterday, 05:18 PM

  (Yesterday, 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 asked GAI and ChatGPT to correct each other's output, and they eventually converged to the following version.  I have removed the "-" to make the eventual comparison to the VMS more fair.  Unmarked syllables are tone 1, all the other tones should be explicitly marked 2-5.
  
  

  Code:

  (A) Kai2 thuek4 daeng: [gae2] ra du khao5 daeng, bam rung sang khan5, un2 that4 fai, ham3 lueat4, cham ra chit2, dap2 phit4, kan sing ap2 mong khon
(B) Hua5:              [gae2] prap2 phi5 sat2 rai4
(C) Man:               [gae2] hu5 tuek2
(D) Sai5:              [gae2] pa sa5 wa4 mai3 yut2
(E) Phang phuet2 nai:  [gae2] thong4 sia5 thong4 ruan
(F) Khi5 kai2 khao5:   [gae2] rok2 hiu5 nam4, khai2 wat4 nao5 ron4
(G) Khon5 pik2:        [gae2] gae2 lueat4 duean mai3 ok2
(H) Khai2 kai2:        []    dap2 phit4 ron4, phu phong fai, kan chak2; [pra2 sit2] pen khong5 sak2 sit2 dang am phan

  
  Does it make more sense?
  
  All the best, --stolfi

• RE: The 'Chinese' Theory: For and Against

  rikforto > Yesterday, 05:19 PM

  Literary Chinese is not the same language as Mandarin or Cantonese. Google Translate copes better with Literary texts than if you were trying to put Latin into the Portuguese translator because of the logographic nature of the text and the fact that Literary turns of phrase are much more current in China than Latin ones are in countries that speak Romance languages, but the language of the SBJ is emphatically not the language that Google Translate is trained to work with. You cannot rely on it for work with text in Literary Chinese.

• RE: The 'Chinese' Theory: For and Against

  Jorge_Stolfi > 7 hours ago

  (Yesterday, 05:19 PM)rikforto Wrote: You are not allowed to view links. Register or Login to view.Literary Chinese is not the same language as Mandarin or Cantonese. Google Translate copes better with Literary texts than if you were trying to put Latin into the Portuguese translator because of the logographic nature of the text and the fact that Literary turns of phrase are much more current in China than Latin ones are in countries that speak Romance languages, but the language of the SBJ is emphatically not the language that Google Translate is trained to work with. You cannot rely on it for work with text in Literary Chinese.

  
  I know that Google Translate is useless for translating the SBJ. (It translated "Minerals, Low grade" to "inferior department of Jade", and worse things).
  
  I am using Google AI, and now also ChatGPT, to translate the SBJ to English and other languages. Those lalamos claim to base their translations on extensive scholarly literature about old Chinese medical texts, in particular the SBJ and its translations in other languages.  Thus they offer detailed explanations for their translations, including references.  But I have not checked the references myself, and they may well be hallucinating.
  
  JoJo's suggestion of using each system to check the proposals of the other has been quite useful. It corrected several mistakes and bad choices that they had made.  In the end I got a text that was approved by both systems.  Of course that still may be very wrong (like the "consensus" of human experts) but it is the best I can do at the moment.
  
  (The other day someone posted a clip of a VMS drawing where one nymph is pulling the towel or hair of another.  It made me recall a famous Japanese block print (ukyo-e) depicting a brawl in a women's public bath. But I didn't remember the name of the artist, so I asked Google AI.  It promptly gave me two answers, with name, date, everything.  But one of them was a peaceful interior scene from a women's bath house without any brawl.  I struggled to find the other one.  In the end I found that there were three famous paintings of brawls in bath houses, both men's and women's.  Goggle AI had mashed together the names f the three artists into a single non-existent new name.  
  
  Anyway, one of the three was the one I remembered.  Of course it is not relevant to the VMS since it is fairly modern, but You are not allowed to view links. Register or Login to view. for curiosity.)
  
  All the best, while it is still possible, --stolfi

• RE: The 'Chinese' Theory: For and Against

  ReneZ > 6 hours ago

  (Yesterday, 05:18 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Does it make more sense?

  
  Yes, this is largely recognisable. The previous version was not Thai at all.
  The number are confusing and make it hard to guess. Thai tones are numbered 1-4 (with no number for 'neutral'), where the assignments are swapped w.r.t. Mandarin (1,3 -> 3,1 and 2,4 -> 4,2).
  hua5 (=head) would be rising = 4th tone.
  
  Thai has many unseparable polysyllabic word. Some seem to appear here but have simply been split, e.g.
  bam rung = bamrung, mong khon = mongkhon . Both words fit very well in this type of text.

• RE: The 'Chinese' Theory: For and Against

  Jorge_Stolfi > 6 hours ago

  (6 hours ago)ReneZ Wrote: You are not allowed to view links. Register or Login to view.Yes, this is largely recognisable.

  
  Thanks! 
  
  However it seems that line (F) of that revised version not quite correct after all.  In the Chinese version, the material for that sub-entry is 屎白 which is literally "excrement white" but must mean "white part of the bird's droppings".  It seems that the Thai translation  "khi3 kai2 khao5" is a bad literal translation of 屎白 that means "white droppings", which is not the actual meaning.  Prompted, Google AI and ChatGPT admitted the probem and said that to get the proper meaning it should have been "khi3 kai2 suan khao5".  
  
  Does that make sense?
  
  

  Quote:The numbers are confusing and make it hard to guess. Thai tones are numbered 1-4 (with no number for 'neutral'), where the assignments are swapped w.r.t. Mandarin (1,3 -> 3,1 and 2,4 -> 4,2).
  hua5 (=head) would be rising = 4th tone.

  
  I can ask the lalamos to change the tone marking system.  The numbers 1-5 were their initial umprompted choice. I asked for a default tone to be unmarked, to lighten the text, and they chose their tone 1.  Is there a name for your marking system?
  
  

  Quote:Thai has many unseparable polysyllabic word. Some seem to appear here but have simply been split, e.g. bam rung = bamrung, mong khon = mongkhon . Both words fit very well in this type of text.

  
  I see.  I split the syllables for the same reason that I omitted all punctuation: to hopefully make it easier to compare this "Thai" text to the VMS text, and (in case Voynichese is indeed Thai) maybe identify additional cribs.  (Of course there would have been many changes to the language, etc.)  
  
  But it may well be that the joined words in Voynichese were actually polysyllabic words of the original language.
  
  All the best, --stolfi

• RE: The 'Chinese' Theory: For and Against

  ReneZ > 6 hours ago

  (6 hours ago)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.It seems that the Thai translation  "khi3 kai2 khao5" is...

  
  Yes, I was wondering about the white chicken poop, (also because the version I saw had khi5, not khi3).
  Should the very last word be amber? amphan is another unseparable bisyllabic word.
  
  W.r.t. your other question, centuries old books exist - I have seen some, but I doubt they go back to the early 1400's.
  I am not sure about that though.

• RE: The 'Chinese' Theory: For and Against

  Jorge_Stolfi > 1 minute ago

  (6 hours ago)ReneZ Wrote: You are not allowed to view links. Register or Login to view. also because the version I saw had khi5, not khi3.

  
  The last round of interactions, pitting ChatGPT agaist Google AI, concluded with fixing a few tones. I don't remember which, but that may have been one of them.
  
  

  Quote:Should the very last word be amber? amphan is another unseparable bisyllabic word.

  
  Yes.  Translations of that part of the Chinese recipe vary a bit, but they generally agree that it is an alchemical use of the eggs, rather than a medical use; specifically, that the eggs can be turned into an amber-like substance.  
  
  All the best, --stolfi
• Next Oldest Next Newest