The Voynich Ninja

I think the crux of the immediate disagreement we are having is that I believe if you propose a spelling system, you should be able to describe the spelling rules to a reasonable degree of accuracy while you do not. Likewise, I do not believe loose correlations between parts of a subset of paragraphs constitute a translation, let alone of the "entire" section, while you do. I will leave it to people reading this thread to judge which of us has the more convincing position

(16-04-2026, 01:46 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.You know, a script vs a cipher thing is one of my "fixations" that I developed while working with Rohonc Codex 

Yes, I am constantly thinking of the Rohonc example.  The proposed origin situations are similar in some aspects -- at least to the extent that in both there was the goal to encode a language that could not be adequately written with the Latin alphabet.

Quote:But cipher doesn't have to be complicated. A cipher may be simple substitution, a sign for a letter. In such case when you memorize the signs you can write with the same fluency as when writing in standard script.

Sure, but very early on it was determined that Voynichese is not a simple substitution cipher for any "European" language.  

There were three possible conclusions from that fact: it is a complicated cipher, it is meaningless gibberish, or the language is not "European".  For some reason, everybody rejected the last one...

Quote:Why Linear B, Cherokee Syllabary or even Rohonc Codex (if I am right) aren't ciphers? Because just like you have a first language so you have a first script and it was the first script for their users.  And If I understand the Chinese theory correctly, Voynichese wasn't a first script for anyone there.

But there are plenty of examples of new scripts developed for languages that already had a "first script".  The modern Vietnamese script replaced a script based on Chinese-like word signs. Same for Korean hangul and Japanese hiragana (although the latter just complemented the Chinese signs, did not replace them).  Or the Shavian alphabet.  A verson of Cyrillic replaced the Mongolian script for the Mongolian language. In the early 1900s Turkey junked their Arabic-like script for one based on a modified Latin alphabet. 

Quote: As I understand in this theory some European traveller wanted to have a copy of some Chinese treatise. Somehow he didn't chose the most obvious options:
- he didn't write it down translated to his own language

 
Because he could not read the Chinese characters; and even if he did, he would not know the names for most of the remedies and diseases in his own language.

Quote:he didn't write it with original Asian letters

For the same reason that he did not just bought a copy of the book. Because such a copy would be totally useless, to him or anyone else, when he got home.

Quote:he didn't write it transliterated with Latin letters

Because there were tones and strange sounds that he presumably felt would be too complicated to encode, and/or because a Latin-based script would take too long to write under dictation.  Which may mean that Voynichese is a bit more a stenography system than a "normal" script.

Quote:People are usually "lazy" and do their job with as little effort as possible. Here he seems to break this rule.

On the contrary, I think he chose the most efficient way he could think of to achieve his goal.

Quote:Some explanation could be that he wanted to hide the meaning.

I definitely reject that idea.  On the contrary, I am sure that he developed the new script (like others developed  pinyin, hangul, etc.) as a way to make the language more accessible, to himself and presumably to others.

By the way, strictly speaking the "Chinese Origin" theory that I have been proposing since the Mailing List days (traveler in "China", dictation by a native, European scribe working form draft, etc.) is distinct from the claim "SPS=SBJ".  The latter has the evidence I found and presented last month. The former is just the hypothesis that I find most plausible given the belief (now fact) that the language is East Asian in a phonetic encoding.

All the best, --stolfi

As far as I understand your theory now, you are claiming that
- some european guy noted one of the most complicated languages
- somehow phonetically 
- in an own system of not more than 20+ characters,
- using at least some of them as „shorthand“ symbols
- and made lots of mistakes and misunderstandings with it.

Even if it is not (central) Chinese but „just“ another far-east language that was based upon chinese writing symbols: 
how is that even possible?

(To be open: I think the funny chinese symbols are already „shorthand“ for chinese language; and they use up to 30,000 of them… Chinese culture was well-developed already ~1,000 years ago — there was no simple-type chinese or other eastasian language, related to Chinese, during our 15th century anymore).

Yes, the mongolians reach deep into Europe from 13th to 16th century, left their marks and used at least 2 different writing systems in that time.
I agree there was very much non-European in Europe and like to emphasize that Europe is not at all limited to an 80km-perimeter around the Alps;
but I don‘t get the „chinese-recording“ traveller with that unique system (written flawlessly by at least 2 different scribers).

Is VMS a book about Europeans? — Yes.
Is VMS a European book? — With a very high probability, yes.
Was it made upon a european language base? We have no idea, but both previous lines point to that.

(16-04-2026, 09:13 PM)Stefan Wirtz_2 Wrote: You are not allowed to view links. Register or Login to view.As far as I understand your theory now, you are claiming that
- some european guy noted one of the most complicated languages
- somehow phonetically 
- in an own system of not more than 20+ characters,
- using at least some of them as „shorthand“ symbols
[...] how is that even possible?

You are not allowed to view links. Register or Login to view. is an alternative to "communist" pinyin that used to be favored in Taiwan and perhaps Hong Kong.  It encodes Mandarin phonetically, including tones, using only 24 of the the 26 Latin letters, without diacritics or numeric indices.

Pinyin, that uses Latin letters plus 4 diacritics to write Mandarin phonetically, was invented in the mid 1500s by an "European guy" who spent several decades in China.  Who not only mastered the spoken language, but also the written one, and translated Euclid and other European books into it.

And you have heard of Marco Polo, haven't you?

Quote:- and made lots of mistakes and misunderstandings with it.

Unfortunately, that is what it seems.  And more errors were added by the Scribe and the BEEP. 

Any phonetic transcription is bound to have many "errors", because phonemes in spoken language are often indistinct.  There is a famous technical paper on speech recognition by computers that is titled "How to Wreck a Nice Beach".

Quote:Even if it is not (central) Chinese but „just“ another far-east language that was based upon chinese writing symbols: (To be open: I think the funny chinese symbols are already „shorthand“ for chinese language; and they use up to 30,000 of them…

Actually one needs to learn only 5000 or so to be "literate". That is about what Chinese students will have learned by the end of high school.

And they are not "shorthand".  Each character stands for one specific syllable, that (unlike syllables of "European" languages) has a general meaning on its own.  These syllables are often joined into compounds of two or more syllables that have specific meanings, loosely related to the meanings of the individual syllables. Like (石 shí "stone") + (脂 zhī "fat") = (石脂 shí zhī "clay").  Or like English "tipewriter": its meaning has to do with the meaning of "type" (as in typography) and the meaning of "writer", but it cannot be guessed from the two parts alone.  A large fraction of the words of European languages translate into such compounds of Chinese.

Quote:Chinese culture was well-developed already ~1,000 years ago — there was no simple-type chinese or other eastasian language, related to Chinese, during our 15th century anymore).

Sorry, I don't understand this comment.

Even today "Chinese" is some 50 different languages, generally not mutually intelligible.  But their syntax is similar enough that they use the same writing system. So the same written text can be read in any of those languages, although the characters are pronounced differently.  There are some differences in syntax and vocabulary, but they seem to be small enough for that system to work.

The Shennong Bencaojing was created around 300 BCE, so it is about 200 years older than the  Disocorides's Materia Medica, and comparable in terms of contents and prestige.  The original text was lost some time before 500 CE, and was then reconstructed from fragments and quotes in other medical books.  

By 1400 CE the reconstructed text was fairly established, and was well-known in the whole area of Chinese cultural influence, including places that spoke monosyllabic but not "Chinese" languages -- such as Vietnam, Cambodia, Laos, Korea, Burma (maybe) and Tibet.   And in Japan too, even though the language there was not monosyllabic. Although in practice it seems that doctors used more extensive and modern reference books. 

It is known that the spoken languages of China changed a lot since the 1400s. For one thing, words that poetry manuals said were rhymes do not rhyme any more.  The evolution of many European languages slowed down considerably after more or less phonetic writing system for them became widely used, since the written literature served as an anchor for the pronunciation.  Italian schoolchildren still study Dante's Divina Commedia (~1200s) as a model of good (if archaic) writing.  But the Chinese writing system, being highly non-phonetic, did not have that effect.

Quote:but I don‘t get the „chinese-recording“ traveller with that unique system (written flawlessly by at least 2 different scribers).

By "flawlessly" you mean "without visible corrections", right?  We can discuss that issue on the "Ink" thread.  Here I would say only that there are in fact many obvious scribal errors which were not corrected. And there may have been many more that were corrected without scraping the vellum.

As for the "2  different scribes", that is still not certain at all.  The illustrations strongly seem to be all in the same hand, and it seems that they were created together with the text.  But at some time some BEEP BEEPed large parts of the text, and it seems that those BEEPed glyphs are mistaken for changes of handwriting. 

Quote:Is VMS a book about Europeans? — Yes.

Well, the Starred Parags section is definitely not about European people, places, or events.  

There is no evidence that the other sections are about "European" stuff.  The illustrations were probably done by an European artist/scribe, but much of their apparent "Eurpeanness" is due to him having copied pictures from European books (often without understanding them).

Even the Zodiac diagrams are rather alien, in spite of the icons at the center.

All the best, --stolfi

(16-04-2026, 11:02 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.You are not allowed to view links. Register or Login to view. is an alternative to "communist" pinyin that used to be favored in Taiwan and perhaps Hong Kong.  It encodes Mandarin phonetically, including tones, using only 24 of the the 26 Latin letters, without diacritics or numeric indices.
Pinyin, that uses Latin letters plus 4 diacritics to write Mandarin phonetically, was invented in the mid 1500s by an "European guy" who spent several decades in China[..]
And you have heard of Marco Polo, haven't you?

You don't mention here that the Gwoyeu... alternative uses Latin characters, but is very frequently doubling vowels and consonants to transport the special emphasizing details of Chinese.
And, at least modern Pinyin uses "only" 25 Latin characters, yes -- but is enhancing letters with diacritics (which is multiplying some characters by x5) and inserts apostrophs, so enhancing the used character set by far.

Nearly none of the VMS characters show anything that could be understood as diacritics or something similiar -- since Voynichese seems to be free of any interpunctation or even anything like "i dots", no wonder.

So your claim is that somebody produced his own "private pinyin" ~150 years earlier before the today known first try? (Ricci)

Marco Polo is a nice-to-have, but I did not hear about any tries of Chinese transscriptions from him, so maybe this is just name-dropping.

[..]

(16-04-2026, 11:02 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.

Quote:Chinese culture was well-developed already ~1,000 years ago — there was no simple-type chinese or other eastasian language, related to Chinese, during our 15th century anymore).

Sorry, I don't understand this comment.
[..]

My line meant that Chinese language(s) were not significantly simpler or somehow "more primitive" 600 years ago.
It was very developed, but rather different from today's versions.
You said this in your words as a reply, so there is no substantial disagreeing.

(16-04-2026, 11:02 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.By "flawlessly" you mean "without visible corrections", right?  We can discuss that issue on the "Ink" thread. [..] glyphs are mistaken for changes of handwriting. 

ok, this would be a complete different discussion somewhere else.

(16-04-2026, 11:02 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.

Quote:Is VMS a book about Europeans? — Yes.

Well, the Starred Parags section is definitely not about European people, places, or events.

You can't say this. 
How can you "definitely" exclude any european content from any paragraphs?
Even if no Europeans are shown, this is not anyhow a proof for texts being "of non-european content". 
Only a valid translation would proove anything there.

(16-04-2026, 11:02 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.There is no evidence that the other sections are about "European" stuff.  The illustrations were probably done by an European artist/scribe, but much of their apparent "Eurpeanness" is due to him having copied pictures from European books (often without understanding them).
Even the Zodiac diagrams are rather alien, in spite of the icons at the center.
[..]

I am holding to what I would call "regionality" or "localism". There is a scientific expression for what I mean with it, but I don't remember or care.
"Regionality" is supposed to mean the relation of motives to their painters/draftsmen and intended viewers: any figures (like Jesus, Angels, Saints, pagan Gods, Celebrities or even normal people) are nearly always presented within the "own" ethnicity in a very lot of medieval artwork.
Depicted figures look very often like their cultural environment, by hair, colour, clothing, time period, weapons & tools etc. -- it is rather rare to find artworks in a "documentation style", showing realistic or naturalistic human beings or animals or landscape.
= 500+ quite european VMS figures, whatever they are intended to show, bring a quite high probability that this manuscript was made also by and for Europeans; somewhere.
But this is just a good chance, not a 100% "proof" of course. I did not claim that.

As a funny sideshow, the Zodiac icons you mentioned could be good enough to turn it to the opposite:
some icon symbols (like the "green dragon" or saurus or whatever, instead of a scorpion) seem to be more a hint to some non-(western)european context than the "copied-looking" Gemini and Sagittarius...

(16-04-2026, 11:02 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.There is no evidence that the other sections are about "European" stuff.  The illustrations were probably done by an European artist/scribe, but much of their apparent "Eurpeanness" is due to him having copied pictures from European books (often without understanding them).

Even the Zodiac diagrams are rather alien, in spite of the icons at the center.

No, the drawings are certainly evidence for European subject matter, albeit not evidence the subject matter is certainly European. The idea that the content of the drawings is a line of evidence for the subject matter of the text is so straightforwardly obvious that I am struggling to articulate how natural it is. I grant, it is not definitive, and there is no need to rehearse to me alternate interpretations or why we should be cautious making sweeping claims with it. But European dress, European figures, European architecture, Western Zodiac, and European media point a certain way and you are the one dismissing evidence out of hand because it contradicts your assumptions about the text

(17-04-2026, 02:10 PM)Stefan Wirtz_2 Wrote: You are not allowed to view links. Register or Login to view.You don't mention here that the Gwoyeu... alternative uses Latin characters, but is very frequently doubling vowels and consonants to transport the special emphasizing details of Chinese.

You mean, like Voynichese has double and triple es, hs, and is?

Quote:Voynichese seems to be free of any interpunctation

This is what the Shennong Bencaojing "rooster" entry would have looked in the original:

   丹雄鸡味甘微温主治女子崩
   中漏下赤白沃补虚温中止血
   通神杀毒辟不祥头主杀鬼肪
   主治耳聋肠主治遗溺肶胵裹
   黄皮主治泄利屎白主治消渴
   伤寒寒热翮羽主下血闭鸡子
   除热火疮痫痓可作 虎魄神
   物鸡白蠹肥猪生平泽

Except that each row would be a column, and the columns would be ordered from right to left.   So the Dictator would certainly not say "丹雄鸡, colon, 味甘, comma, 微 period".  And so the Author would have hardly known where to insert punctuation on his own.

Quote:or even anything like "i dots"

And that is part of the reason why he did not use Latin letters: they are too slow to write.  In his alphabet, most glyphs (even t, k, and Ch) require only one or two strokes simple pen strokes.  Sh requires 3 because of the diacritic plume, and CTh etc require 4.

Quote:So your claim is that somebody produced his own "private pinyin" ~150 years earlier before the today known first try? (Ricci)

Yes. For the same reason that Ricci created his proto-pinyin: because he was not fluent in the Chinese writing, and knew that anything written that way would be inaccessible to his intended readers.

The creation of pinyin (or the Vietnamese script) did not depend on any prior development.  Pinyin could have been invented by a Phoenician 3000 years ago. The only thing that Ricci needed was the motive.

Quote:Marco Polo is a nice-to-have, but I did not hear about any tries of Chinese transcriptions from him, so maybe this is just name-dropping.

He is just the most famous example of hundreds (if not tens of thousands) of Europeans who traveled to "China"  and lived there long enough to be in the situation proposed by the Chinese Origin theory.   By "China" of course I mean any East Asian country with a monosyllabic language.  Marco may have been too occupied with his jobs at the Court, and probably also with Chinese girls (he was 21 when he got there) to undertake the "VMS project".  But any other traveler who spent years there around 1400 could have done so.

Quote:

(16-04-2026, 11:02 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Well, the Starred Parags section [SPS] is definitely not about European people, places, or events.

You can't say this.  How can you "definitely" exclude any european content from any paragraphs?

Well, all analysis and statistics I have done so far is consistent with the SPS being nothing more than a transcription of the Shennong Bencaojing (SBJ), in some as yet unidentified language and script.  The SBJ is supposed to have 365 recipes, which is just about the estimated number of SPS parags before the loss of those pages.  The distribution of lengths of the surviving SPS parags matches that of the lengths of SBJ recipes.  Of the 30+ attempts I did so far to find SPS parags matching SBJ recipes, about 20 yielded a parag that clearly matches the recipe; and there is no sign in the parag of any extraneous text.

It is theoretically possible that the Author replaced some SBJ recipes by parags with similar size and structure, of his own or taken from some European source.  Alternative recipes, prayers, gossip about a improper dealings of a certain papal candidate with a secret fraternity of five dirty nuns who fantasized about warm baths.  But I have seen no evidence for that yet...

Quote:any figures (like Jesus, Angels, Saints, pagan Gods, Celebrities or even normal people) are nearly always presented within the "own" ethnicity in a very lot of medieval artwork.
Depicted figures look very often like their cultural environment, by hair, colour, clothing, time period, weapons & tools etc.

Quite true.  This is a copy of a painting by Paul Veronese that my mom did.  Can you tell which Biblical scene is being depicted here, and who that lady is supposed to be?

[attachment=15149]

Quote:500+ quite european VMS figures, whatever they are intended to show, bring a quite high probability that this manuscript was made also by and for Europeans; somewhere.

And that is in fact what the Chinese Origin theory proposes!

All the best, --stolfi

(17-04-2026, 04:12 PM)rikforto Wrote: You are not allowed to view links. Register or Login to view.No, the drawings are certainly evidence for European subject matter, albeit not evidence the subject matter is certainly European. [...] ... European dress, European figures, [...] , and European media point a certain way. [...] The idea that the content of the drawings is a line of evidence for the subject matter of the text is so straightforwardly obvious that I am struggling to articulate how natural it is.

I know, that is a very natural mistake.

Most elements of the drawings are clearly just decoration.  Even if the hairdos and poses of the Zodiac nymphs are encoding some part of the contents, their "European" appearance is certainly not determined by the encoding.   Those are not portraits of specific people.  The Artist could have used winged monkeys, trees, fishes, Inuits in winter clothing with Bantu spears, ... The choice or European-looking ladies with light hair was a decision by the Artist (with or without input from the Author) determined only by the intention of making those diagrams "look nice".

Quote:European Architecture

Same observation.  There is no reason to think that the castles in the big fold-out are faithful depictions of real castles.  The whole archipelago is clearly imaginary or symbolic.  The significant contents almost certainly just "there is a castle guarding/blocking the imaginary bridge between these two imaginary islands."   The shape and appearance of the castle, including the merlons, surely were defined by the Artist -- based on castles in his vicinity, on his memory, or on books that he had at hand.

And in fact there are some architectural details in that fold-out that do not quite look European. There is a minaret tower, a volcano, two large baldachins ... and six onion-roof towers at the center.  Which some desperately try to interpret as giant apothecary jars in order to preserve the "European" claim...

Quote:Western Zodiac

The only "Western" thing in that Zodiac, besides the decorative nymphs, are the depictions of the signs at the center.  Which are decoration, too.  Surely the Author did not put those drawings there to tell the reader "this is what the sign of Scorpio looks like".  

The Zodiac diagrams themselves are rather unusual for European astrology, because each sign i divided into 30 degrees of arc, 360 in total -- instead of 30/31 days, 365 in total.  That conception was not unknown in Europe; but what is (AFAIK) quite unusual is the division of two signs into halves of 15 degrees.  On the other hand, both of those features are familiar in Chinese astrology.

All the best, --stolfi

Quote:Can you tell which Biblical scene is being depicted here, and who that lady is supposed to be?

Baby Moses taken out of the water by Pharaoh's daughter in Egypt? 
(I have learnt to recognize such stuff a bit)

Here is another one:
[attachment=15150]

A quick note:

I know I brought up the cross-check between two LLMs myself. However, I mainly use it as a "numerical" plausibility check – by running the same count or statistical analysis through Claude and ChatGPT to detect hallucinations. It is unlikely that both LLMs would produce the same hallucinations – i.e. the same values – during a count or statistical analysis. This has allowed me to spot many minor errors, with ChatGPT hallucinating significantly more and constantly refusing to actually count, simply spitting out random numbers, compared to Claude. This is also because I have internally prohibited Claude from hallucinating.

It is far more critical to use LLMs for translation.
When it comes to languages, LLMs sometimes have the same training data and will therefore hallucinate similar patterns. So as soon as two LLMs agree on the phrasing based on their training data, I can no longer distinguish between what is a hallucination and what is the truth. Gemini and Google Translate in particular get very creative when they don’t know what they’re dealing with.

That’s why I try to keep this part manual, or check against real dictionaries as soon as a decent level of accuracy is required. Since I actually do this, I’ve also noticed just how much hallucination appears in the translations, particularly with Gemini, but also with ChatGPT and even Claude.

As I said, just a note.