The 'Chinese' Theory: For and Against

[Jorge_Stolfi]

Scribe's mistakes, lunar month, phonetic... These are the assumptions!  ... An assumption is something you cannot prove, but you take it as truth. ... you haven’t been able to prove anything yet

There is no such thing as real "proof" outside of mathematics.  There are only probabilities.  In real life, we say that a thing is "proved" when our probabilty for the alternatives is too small for us to bother with them.

(And note that I said "our" probability. Because there is no such thing as "the" probability, either.  The probability of a statement is a numeric expression of our belief is true. Therefore "my" probability is different from "your" probability which is different from that other guy's probability.  Because the probability is determined by the totality of what one knows and how one reasons.)

All those "assumptions" you cite are deductions based on what I have seen in the SPS, even before thought of comparing the SBJ with the SPS.

Scribe's mistakes

 

It is impossible to avoid that conclusion if you try transcribing any significant part of the VMS from the scanned images (as I did -- 80% of the text since last year, not counting what I transcribed back in the 1990s).  

More objectively, there is the large number of "weirdo" characters and of words that violate the structural models, seemingly in a random way.   There are also the misaligned lines on f34r, the missing arms and stars in the Zodiac, the dozens of parags mashed together in the SPS,  and more. 

The Scribe was obviously very careless and hurried in many places. The apparent absence of corrections also indicates that the text was not proof-read (like other manuscripts of the time were).

Based on those observations, I had estimated that about 5% of the tokens had an error of some sort.  Before I started trying to match the SBJ and the SPS.

Again, that conclusion naturally is unacceptable to those who believe that the VMS is encrypted.  But those who think it is gibberish, or Greek, Slavic, Bavarian etc should have no problem with it.

lunar month

I never claimed there was some "lunar month" stuff in the VMS.  On the contrary, that concept is notable for its absence.  Instead, what I see is that the Zodiac section divides the year into 12 sections of exactly 30 "things" (or 24 sections of exactly 15 "things").  That is neither the Western Zodiac nor a solar or lunar calendar.   

I found one system that divides the Ecliptic into 24 sections of exactly 15 "things".  The problem is, it is a Chinese system called the "solar terms" ...

phonetic

As I explained back in 2002, I was forced to accept the theory that Voynichese was a phonetic rendering of a monosyllabic language because of the results I got while trying to prove that it was not natural language. 

As you may have seen, at first I though of another possibility: that it was a codebook cipher using some funky Roman-like number system. But that was very unlikely because such ciphers are too hard to encode and decode for a document of that size.  I switched to the "Chinese" (East Asian) theory only after I found that phonetic Vietnamese and Mandarin gave the same "unnatural" statistics as Voynichese.

And then there are the arguments against encryption in general that I states in a previous post.

That was well before I knew of the Shennong Bencao.  The Chinese Dictaton scenario too came later: it was a proposed explanation of how a phonetic East Asian text would have been written down by an European Scribe on European vellum.

you call the theory “the only one of its kind” is not true—it’s fundamentally wrong.

I did not say that it was "the only one of its kind".  I said that it is the only one I know that can plausibly explain why the VMS looks that way and why it was created.

In fact I claim that it is the only theory that explains all the puzzling features of the VMS -- as natural consequences of how it was created; which in turn was a natural consequence of why it was created.

I can easily explain why this book was written, and the explanation is as simple as it gets and raises no questions: it’s a medical reference book. That’s it.

More precisely, it looks like a collection of ~4 separate books that include a herbal, a materia medica, an astronomical and astrological treatise, and a treatise that may be about anatomy of about medicinal baths.  

But that "explanation" alone does not explain why the Author had to devise a new alphabet, why the glyph look like that, why it had to be encrypted but at the same time it had to be illustrated with bizarre diagrams and "bathing" scenes.  It does not explain the complex word structure; the absence of identifiable articles, pronouns, copulas, etc.; the repeated words; and a lot more.  

If you didn’t understand me, I’ll explain it to you clearly. You think that daiin-kaiin-laiin-dain are equivalent spellings of the same word ... This is an assumption

Call it how you wish, but that was a deduction that followed from comparing the longest SBJ recipe ("red rooster") to the longest single-star parag of the SPS (f105v.32).  Again, that recipe has eight 主 characters, and at the corresponding positions of that parag one finds five daiin, one dair, one laiin, and one kaiin. All within a few EVA letters away from the expected positions. Hence the conclusion that laiin and kaiin were alternative "spelling" of daiin as the Voynichese equivalent of 主.

As for dair, the proposal that "-ir" was an alternative form of "-iin" was made earlier independently of the SPS=SBJ theory or any assignment of meanings to the glyphs, phonetic or otherwise.  It was justified by the analysis of the word structure and the frequencies of the "coda" elements in that structure.  If that identification is accepted, then six of the eight 主 of "red rooster" have daiin in matching positions of that parag.

We cannot know why these errors were made (I have already asked this question)

And I have answered it.

As explained above, the claim that daiin is sometimes replaced by laiin or kaiin is simply an observation from the data.  

Why and how those replacements happened is indeed open to speculation.

One possible explanation is indeed that the Scribe misread the Author's ds as ks, ls, or maybe other letters. As I pointed out, the k is sometimes written by the scribe as a short letter, only about as tall as a d.  If that is how the Author wrote his ks in the draft, and his handwriting was a bad as mine, the Scribe's confusion would be quite understandable.  On the other hand, the Scribe would hardly ever confuse a t with a d, because of the double loop.  And indeed I have not yet found a recipe which requires matching a 主 to a taiin.

Another possible explanation is that the Author had some difficulty distinguishing the sounds of k, d, and l as pronounced by the Dictator.

Yet another possible explanation is that the sound of the d in daiin changes depending of the previous word.  Like in English the article "a" changes to "an" before a word that begins with a vowel, or Japanese changes the syllable "hi" to "bi" or "pi" depending on the previous syllable (100 = "hiaku", 200 = "nihiaku", 300 = "sanbiaku").

Which explanation applied in each case is not important.  What matters is that there are plausible explanations for those variations.

the foolish scribe (where did the author even find him?)

That the Scribe was sloppy is also an observation from the text and drawings of the whole book.

It has long been observed that the quality of the vellum is very poor.  The natural deduction is that the Author was financially strained when he set out to put his theories or whatever onto vellum.  Then it is not surprising that he recruited such a bad scribe, and had to put up with his errors. 

Also, there were no cataract operations at that time.  Thus people often became functionally blind when they got old (like I would be if I had not operated both eyes long ago).  With failing eyesight, they would be unable to read and write on their own, and would have to rely on secretaries or public writers (which seems t have been the case of Marci by the time he sent the book to Kircher).  They would also be unable to check the work of those scribes for errors...

[are] such results are applicable to other parts of the manuscript, etc.

The statistics of Voynichese seem to be pretty uniform across all sections.  Even the language A/B differences are small relative to their similarities.  In particular, weirdos and malformed words are roughly equally common in all sections.  Thus it is indeed likely that 5% or more of the tokens are variations from a "canonical" word.  In particular, it seems likely that, in all sections, many instances of daiin were replaced by kaiin or other variants, and may -iin endings became -ir.

 the fact that one recipe works doesn’t say anything. Absolutely nothing.

That is silly. YOU may say that those coincidences that I pointed out are not enough to make you believe in the SPS=SBJ claim.  But cryptographers would jump on their seats if they found much less than that in their puzzles.

All the best, --stolfi