(13-02-2026, 02:10 PM)Yavernoxia Wrote: You are not allowed to view links. Register or Login to view.IMHO, the core scenario of the backstory, namely a European learning spoken Chinese
Have you heard of Marco Polo?
Marco is only the most famous one, but we know of dozens of other people who traveled to China and other East Asian countries, lived there for a few years, and learned the spoken language. And if there are dozens of
known cases, there must have been hundreds if not thousands of people who did that and were alive in the 1400s.
Quote:inventing a phonetic script
Again there are dozens of examples of Europeans inventing phonetic scripts for other languages. Pinyin and the modern Vietnamese script were invented by European Jesuits (in the late 1500s and in the 1700s, respectively), for languages that already had their own native scripts.
Quote:and having a scribe
I believe it was the Author himself, not a scribe, who took dictation. He would have recruited a Scribe later, to put his notes into vellum.
Quote:transcribe a Chinese text orally into an invented phonetic script
Every linguist studying a "new" language does that.
Quote:is extremely far-fetched.
I myself once helped a desperate Linguistics student in the US to debug and process the digital copy of a file-cabinet-size punched card copy of a typewritten copy of a pile of notes dictated in the 1800s to a Bureau of Indian Affairs officer by some Native Americans whose tribe and language are now extinct. In a spelling system devised by that officer for the purpose. Luckily for that student, I had used punched cards in my graduate course, and so I could tell him how to correct those errors that resulted from the card reader missing a hole in a punched card.
Is that "extremely far-fetched" too?
Quote:If the author already understood spoken Chinese well enough, he could simply have translated the text into his own language instead of inventing a new script.
See my reply to @MHTamdgidi_(Behrooz) about that.
Quote:relying on oral dictation through intermediaries would have produced huge errors
Yes! And that is one fact that all statistical and structural analysis must take into account.
The text is full of errors.
How many? Hard to tell, but I would guess at least 1-2 glyphs per line. Maybe more. Not just from dictation, but from being copied to vellum by an ignorant Scribe, then BEEEP, then transcribed digitally by humans
Quote:yet the text shows patterns that are far too regular to be explained as random dictation artifacts
Quite the opposite! There are literally thousands of irregularities for which "error" is the most likely explanation. Like the several hundred "weirdo" glyphs that occur only once: many of them must be just sloppy or garbled versions of ordinary glyphs. Or the many words that violate the apparent structural model, like words with two gallows, or "forbidden" combinations like
se and
qd. And the hiccups in the 4x17 sequence of You are not allowed to view links.
Register or
Login to view. and the 3 x 7 sequence of f49v. And much, much more.
Quote:The claim that the European Origin theory is supported by 'ABSOLUTELY ZERO evidence' is scientifically untenable. It ignores the entire forensic reality of the object. Let’s look at the '408 tons' of hard evidence that anchor this object to Europe, specifically Northern Italy:
- The radiocarbon dating (1404–1438) places the vellum in the early 15th century.
That tells us absolutely nothing about the language, or where it was written. Or even
when it was written.
Quote:The ink chemistry (Iron Gall) is consistent with European recipes of that period. The quire binding is European. the physical artifact is European. "The analysis of castle and dress drawings... makes it almost certain that the Scribe was from Northern Italy."
And you can add to that the writing instrument (quill), the layout of the text (left to right, split into parags, with the tail line of each parag left-justfied), the cloud bands, the T-O maps, the "four stages of life" on f85r2, the root dragons in the Herbal, and a lot more.
Quote:If the scribe is Italian, the materials are European, and the style is Venetian/Lombardic, then the text must also be of local origin.
And that is why 100 years of concentrated efforts by the best paleographers and cryptographers failed to decipher a single label. A single glyph.
Because
everybody who had the least chance of doing so -- meaning anyone who knew what a medieval manuscript looked like --
made that same stupid logical mistake.
All that "evidence" may make it likely that the
Scribe was European.
But that is it. All those drawing elements that look "European" are
decoration. Surely the author did not specify how each nymph had to be dressed and how she would hold her hands. It makes no sense that he would have wanted nymphs so that he could encode some astrological information in those details. He wanted them only because he wanted the zodiac diagrams to look pretty, like those that Europeans were used to. The real information contents of each Zodiac diagram is only the rings of text, the list of 30 (or 15) labels, and the approximate time of the year that the diagram refers to. So he told the Scribe to draw a nymph and a star for each label, arranged in a circle -- like most every European book would present astrological information.
The conclusion that "If those things look European, then the language/contents must be European"
is a gross logical fallacy. A non-sequitur. The stated consequence does not follow at all from the premise. You can't put a "therefore" there.
Quote:The 'bench' or 'gallows' are common in legal shorthand of the era.
You cannot write a symbol with two simple pen strokes without it looking just like some symbol from twenty different manuscripts.
The suggestion that the gallows may be scribal abbreviations makes no sense. A system of abbreviations would have at least a dozen such abbreviations in addition to the letters of the alphabet. But there are only two "everyday" gallows -- t and k -- and two "fancy" ones -- p and f .
Quote:It is far more logical that a European scribe used familiar abbreviation symbols to write a constrained European code
I agree: If the scribe is European and the language is European, then he would understand it, and he would know when to use abbreviations, and then it would be more logical for him to write in an abbreviated European language.
But there seems to be a gap somewhere with this argument too. Hmm..
Quote:a new phonetic script that coincidentally looks exactly like a Latin chancery ledger.
Voynichese looks exactly like a Latin ledger only in the same sense that the cuneiform script looks exactly like bird footprints on wet sand. That is: only for the first three seconds when you first see it.
Quote:You cite low entropy as proof of a monosyllabic Asian language.
You got it wrong. I don't say such thing. Many say that the
low per-character entropy is proof that it is
not a simple substitution cipher of an European language, or even a natural language. That again is a non-sequitur. The per-character entropy cannot be used to identify the language, or distinguish natural language from gibberish.
What I have noted many times is that the
per-word entropy (about 10 bits) is compatible with many natural languages,
including East Asian ones, under any spelling system or encoding where each word is (almost) always spelled/encoded in the same way. In fact, IIRC, it was rene who first made this observation.
[quote]the null hypothesis must be that the text itself is also of local (European) origin.[quote]
You are misusing the word "null" there. It is used when testing drugs (foods, etc.) There there are two alternatives: the drug has some perceptible effect, or it has no perceptible effect. The latter is called the "null" hypothesis for the obvious reason. For the drug to be approved, the results of the tests must be such that they make the null hypothesis sufficiently unlikely. That is, the effects of the drug observed in the tests must be large enough that the probability that they are due to random experimental noise is very small. Note that the "null hypothesis" does not mean "the most likely hypothesis".
That scenario does not apply in the case of Voynich Manuscript. The "European Language" (or "European Contents") is not a "null" hypothesis in any sense. It is just
one of the alternative hypotheses for the language, like the "East Asian Language" hypothesis, the "Semitic Language" hypothesis, and "African", "Amerind", "Austronesian" etc "Language" hypotheses.
Now, each one of us has the right to assign "a priori" probabilities to each of those hypotheses. If yu gave me a manuscript book and told me only that it was written in the 1400s and in the 1600 it was recorded as being in Prague, before opening that book my probabilities for its language would be something like
- Latin 60%
- German 15%
- French 7%
- Hebrew 5%
- Spanish 3%
- Greek 3%
- Arabic 2%
- Czech 2%
- Gibberish created mechanically to fool a gullible Emperor 0.01%
- Invented language used by a small community of lesbian nuns 0.0001%
- Other 2.9899%
There may be statistics about that question, and they may be very different from the numbers above; but I don't know any, so I must pick my probabilities based only on what I know of history etc.
But if then I open the book and see a script that does not resemble any that I know, not even Rongorongo or the Phaistos glyphs. Then my probability of "Other" increases considerably, at the expense of the others. And my probabilities will keep changing the more I study the thing.
Probability theory says how you should modify your prior probabilities as you obtain more information. Basically, you ask for each hypothesis Hi and each consequence Cj, "if Hi was true, what is the probability that we would observe Cj". If, under some hypothesis H7, the probability of C5 would be very high, but you don't see C5, that would push your probability of H7 down. Conversely, if you do see C5, that will push the probability of H7 up. There is a formula for that, but you should get the idea.
Well, if the "European Language" hypothesis was true, with high probability the VMS would have been deciphered by now. Or at least we would have some understanding of the type of encryption. And we would have been able to identify patters characteristic of European languages, such as articles and prepositions, the verb/adverb/noun/adjective partition of the lexicon that is a mark of Indo-European languages, etc. And the word lengths would have a long-tailed distribution. Etc.
But none of those expected consequences has happened.
So, rationally, the "European Language" hypothesis should now have very low probability.
On the other hand, the East Asian Language hypothesis predicts certain thing about the language, such as rigid word structure, limited word length, and absence of any perceptible grammar, and that it would have many features of natural languages like 10 bits of word entropy, Zipf's law, and correlations of word use with subject matter and between adjacent pages. And it predicts that attempt to decipher it by starting from the assumption that it must be an encrypted European language will completely fail. And all these predictions have been observed in Voynichese.
But that was before the discovery that the SPS is a close, almost word-for-word version of the Shennong Bencaojing. Now you may use the term "null hypothesis" for the alternative to that claim: that the SPS is
not the SBJ, but an unrelated book -- original or copied/translated from some other book unrelated to the SBJ.
So now, to prove that claim SPS=SBJ, we would have to show that some consequences predicted by this hypothesis are unlikely to be the result of random chance, as they would have to be under the SPS≠SBJ hypothesis.
In particular the SPS=SBJ hypothesis predicts that the longest paragraph of the SPS should have multiple repetitions of the word daiin, the most common word in the SPS, at specific intervals that match the intervals between the occurrences of 主, the most common character in the SBJ, in the longest SBJ recipe. And that is what we see. See my reply to @dashtodfk for the computation of the probability of that happening by chance, under the SPS≠SBJ hypothesis.
All the best, --stolfi
