The Voynich Ninja

Depending on what you mean with layman, but in my understanding of the 15th century society, the creator of the manuscript writing system had to have some qualifications and knowledge that were not at least so common in that era.

The creator of the manuscript's writing system must have been well educated in Latin to scribal proficiency and schooled in its inflectional grammar, such that the stem-and-ending structure of the word functioned as an internalized model. Possibly also further trained in the positional numerate tradition of commerce,as I believe these two formations, grammatical and arithmetical, underlie the writing system's architecture.

He also has to have had exposure to at least one, but possibly more, non-Latin scripts and languages like hebrew ( final-form allography, root-and-pattern word-model), sufficient to derive a structural features from them and to apprehend the separability of script from language, but also the exotic-alphabet compendium genre at a depth adequate for convincing imitations.

He possessed some knowledge of an astro-medical practice — herbal, pharmaceutical, calendrical, balneological, and gynecological — at the taxonomic level required to enumerate and factor the domain, though not necessarily at the level of its foremost practitioners.

All this requires no university formation or no cryptographic training but at least personal interest to it, and no mathematics beyond enumeration.
It might indicate an unusual conjunction of otherwise ordinary competences in one individual.

In my understanding, no one in that era had applied a positional slot-system to the encoding of text.
The positional micro-notations were used, but every one of them served number, record, or shorthand; the transfer to running text has no living precedent within his reach.

My conjecture: To encode their original source text, before the creator designed the glyphs, he did estimate how many glyphs were needed to produce the desired result.
How he estimated is an interesting exercise of thought. He had exposure to other languages, and some of them have around 23 letters like Greek, Sami, native Italian, classical Latin or standard Portugese, and Hebrew.
The rigid positional system he designed further restricted the expressivity, so he must have taken that in account too. To write a 1000 word source document, what kind of system I need, how many glyphs, how many positions, how many rules/grammar.

Another conjucture: What he did is possibly a lossy encoder, not encoding everything, but an intended receiver is able to recover the missing information - from the context or other clues (markers, flags), or has acces to other material ("reading guide")

(15-07-2026, 08:32 AM)Pointless.. Wrote: You are not allowed to view links. Register or Login to view.Depending on what you mean with layman, but in my understanding of the 15th century society, the creator of the manuscript writing system had to have some qualifications and knowledge that were not at least so common in that era.

Well, yes.  But the required knowledge is not that great, and many middle-class people would qualify.  Like Marco Polo and his uncles.

Quote:The creator of the manuscript's writing system must have been well educated in Latin to scribal proficiency

There are bits of evidence and common-sense arguments saying that the Author who composed the text and conceived the basic design of the diagrams was not the same person as the Scribe who actually put pen to vellum.  So we have no information about the scribal abilities of the Author.  

The glyph shapes may resemble those of Latin letters at first sight, but that impression quickly dissipates when one looks at them more closely.  Voynichese glyphs were mostly created by combining a set of half a dozen simple strokes like e, i, ' in pairs in almost all possible ways.  One reason for the superficial resemblance with Latin letters is that both were designed to be written with a quill pen, which led to the simplest strokes having similar shapes.  

Only three letters of Voynichese -- e, a, o -- match letters in the Latin alphabet.  Voynichese r is clearly different from Latin "r", especially in the way the latter was written back then.  Voynichese i is always dotless and slanted NW-SE, whereas Latin "i" is supposed to be upright and often needs the dot.  Voynichese f always has a loop at the top, very unlike Latin "P".

Another reason for the superficial resemblance of Voynichese to Latin script is that both have (1) short "minuscule" symbols, some with ascenders and descenders, and (2) tall "majuscule" symbols with no descenders, resting on the same baseline.  But feature (1) is shared with other scripts like Greek, Armenian, Georgian, Cyrillic,etc; and several of them have also (2).  Anyway, the tall glyphs of Voynichese seem to have very different properties and uses than the majuscules of other languages.

In fact, a notable feature of the Voynichese script is that it has no significant resemblance to any other known script.

Quote:and schooled in its inflectional grammar, such that the stem-and-ending structure of the word functioned as an internalized model. ... He also has to have had exposure to at least one, but possibly more, non-Latin scripts and languages like hebrew ( final-form allography, root-and-pattern word-model), sufficient to derive a structural features from them.

The words of Voynichese have a rather rigid structure, with multiple slots (10 or more) each of which can be left empty or filed with one of a small number of alternatives. 

This structure does not resemble that of words from any "European" language, including Hebrew and Arabic.  All attempts made in the last 100+ years to match those words to those of "European" languages, or to deduce the grammar of Voynichese, have completely failed.  

Those who still believe that Voynichese is an "European" language are forced to admit that the the whole VMS has been encrypted with a complex cipher that completely obscured its original lexicon and grammar.  

Thus, in particular, we have no clue favoring any particular "European" language over any other.  The few parts that we can read -- the folio and quire numbers, the month names, and the "marginalia" -- were most likely added after the Author was no longer around; and the last two seem to be in two different languages.

Quote:Possibly also further trained in the positional numerate tradition of commerce

No one has been able to identify any "words" that could be numbers, either.  So we don't know anything about the Author's mathematical abilities. (But we can see that the Scribe did not know Euclidean geometry at all.)

Quote:He possessed some knowledge of an astro-medical practice — herbal, pharmaceutical, calendrical, balneological, and gynecological

If the contents of the VMS is original material from the Author's mind, yes, he must be a polymath.

But the Author may have simply copied or summarized existing books, translating them to Voynichese.  In that case, he may have known as much about those subjects as the scribes in monasteries that produced all sorts of treatises for sale or commission.  That is, maybe even nothing at all.

Quote:My conjecture: To encode their original source text, before the creator designed the glyphs, he did estimate how many glyphs were needed to produce the desired result.

That seems to be a very safe bet...

Quote:Another conjecture: What he did is possibly a lossy encoder, not encoding everything, but an intended receiver is able to recover the missing information

It is highly likely that he left out some details of the original text as it would have been spoken by someone who knew it and its meaning.  Every script, even the ones that were created or reformed in modern times, is "lossy" in that sense.

At a minimum, he probably did not bother to record pauses and rhythm, including the pauses between sentences.  Few scripts bother to indicate the stressed syllable in a word.  Many use the same letter to indicate two or more very different sounds.  It would be remarkable if Voynichese did not have such "features" too.

All the best, --stolfi

Thanks for the food for thought. It took some careful reading, as my exposure to the Voynich comprendum is still rather limited and many of the concepts you mentioned are, for the time being, somewhat obscure to me.

Quote:There are bits of evidence and common-sense arguments saying that the Author who composed the text and conceived the basic design of the diagrams was not the same person as the Scribe who actually put pen to vellum.  So we have no information about the scribal abilities of the Author. 

If we assume the manuscript emerged from a family or guild environment, it makes sense that its production stayed entirely within that closed circle. The rather naive illustrations actually may support the idea of a small group—probably under twenty people—that simply lacked artistic skills. While they could have hired an outside illustrator, the final artwork would make that an impressive choice. As for the text itself, the writing system was necessarily architected for fluidity, suggesting the creator had a solid grasp of scribal processes. However, I have doubts that the system's inventor actually the one or the few holding the pen.

Quote:The glyph shapes may resemble those of Latin letters at first sight, but that impression quickly dissipates when one looks at them more closely.  Voynichese glyphs were mostly created by combining a set of half a dozen simple strokes like e, i, ' in pairs in almost all possible ways.  One reason for the superficial resemblance with Latin letters is that both were designed to be written with a quill pen, which led to the simplest strokes having similar shapes. 

There seem to be at least two plausible ways of looking at this.

One possibility is that the system was designed from the top down. If the creator was working against the clock, the basic character inventory may have been chosen fairly quickly, leaving more time for the harder task of developing the encoding system and making it structurally flexible. Once the set of symbols was fixed, someone could then have refined their practical execution, simplifying them into the stroke model you describe. That would have helped scribes practise efficiently and develop the speed, fluency, and muscle memory required for continuous writing. So, I think there is a possibility that there was a calligraphic optimization phase made by someone else, after the glyphs were decided.

But the process may equally well have worked in the opposite direction. Perhaps it began with a small set of basic strokes. By experimenting with different combinations, the creator could then have built up the complete vocabulary of symbols from those simple elements.

There is a conjecture, I believe, which says that the physical shape of a symbol could be related to its position within the text. It may have some relation to this, but I have not yet read about it enough to understand it well enough, so I won't discuss it further here.

Quote:Only three letters of Voynichese -- e, a, o -- match letters in the Latin alphabet.  Voynichese r is clearly different from Latin "r", especially in the way the latter was written back then.  Voynichese i is always dotless and slanted NW-SE, whereas Latin "i" is supposed to be upright and often needs the dot.  Voynichese f always has a loop at the top, very unlike Latin "P".

Another reason for the superficial resemblance of Voynichese to Latin script is that both have (1) short "minuscule" symbols, some with ascenders and descenders, and (2) tall "majuscule" symbols with no descenders, resting on the same baseline.  But feature (1) is shared with other scripts like Greek, Armenian, Georgian, Cyrillic,etc; and several of them have also (2).  Anyway, the tall glyphs of Voynichese seem to have very different properties and uses than the majuscules of other languages.

In fact, a notable feature of the Voynichese script is that it has no significant resemblance to any other known script.

I won't go too far into the manuscript-specific details, partly because my own understanding of the manuscript and its historical context is still fairly limited. You have much more experience with the subject than I do, so I read your observations carefully.

That said, I can certainly agree that the system as a whole does not match any known script. My impression, however, is that some of the individual glyph forms look at least somewhat similar to forms that appear in various known writing traditions. If the creator consciously designed the system, it seems reasonable to assume they were exposed to more than just Latin script and had at least some familiarity with letterforms from other traditions as well, meaning ultimately that he'd absorbed at least one structural idea from a non-Latin system.

I also agree that the glyph inventory is not a letter-for-letter match to any known alphabet. In fact, that is something that I would expect from a constructed system. If someone was creating a new script, we would anticipate a mixture of borrowed inspiration and original design choices, rather than a direct copy of any existing model.

For that reason, I'm not entirely convinced that a purely comparative description of the glyphs is the right tool for addressing the question. A constructed system is precisely the sort of thing that should fail direct comparisons while still making perfect sense on its own terms. If the system was deliberately invented, then the lack of a close match is not evidence against that idea. If anything, it may be one of the expected outcomes.

Put another way, it seems difficult to detect a deliberate invention simply by comparing it against a catalogue of things that already existed. Invention, by its very nature, is the act of creating something that is not already in the catalogue. To my mind, the more interesting question is whether the glyphs exhibit internal logic and design choices that are consistent with someone intentionally constructing a writing system, rather than whether they can be traced back to a single known source.

Quote:The words of Voynichese have a rather rigid structure, with multiple slots (10 or more) each of which can be left empty or filed with one of a small number of alternatives.

This structure does not resemble that of words from any "European" language, including Hebrew and Arabic.  All attempts made in the last 100+ years to match those words to those of "European" languages, or to deduce the grammar of Voynichese, have completely failed. 

Those who still believe that Voynichese is an "European" language are forced to admit that the the whole VMS has been encrypted with a complex cipher that completely obscured its original lexicon and grammar. 

Thus, in particular, we have no clue favoring any particular "European" language over any other.  The few parts that we can read -- the folio and quire numbers, the month names, and the "marginalia" -- were most likely added after the Author was no longer around; and the last two seem to be in two different languages.

Where I diverge is that I am not entirely convinced we are looking at a language encoded onto the page. If my admittedly speculative "lossy encoder" idea has any merit, then the system may have been designed for a receiver who already possessed substantial background knowledge. In that situation, many language-like features could be omitted quite deliberately. Function words, much of the inflection, and even parts of the grammatical structure might be treated as redundant and therefore left unstated.

What remains would not necessarily be a reduced language. It might not be a language at all, at least not in the linguistic sense. If grammatical information was never recorded, then no amount of analysis can recover it. The linguistic layer would have existed, if anywhere, only in the minds of the writer and the intended reader. It was never actually transferred onto the vellum.

Under that view, the manuscript would contain only the information that the recipient could not reliably reconstruct from existing knowledge. Everything else could remain implicit.

Of course, all of this is conjecture. I can't prove it, and at present it is little more than a working hypothesis. Still, after reading a great deal of Voynich research over the years, especially the extensive statistical analyses, I find myself increasingly inclined toward the possibility that there may simply be no underlying language to recover.

So my current position is perhaps an unusual middle ground: I do not think the manuscript is a hoax, nor do I think it is meaningless. At the same time, I am not convinced that it is fundamentally linguistic. It may instead be encoding something else entirely, with language serving only as an unrecorded layer shared between the sender and the receiver.

Quote:No one has been able to identify any "words" that could be numbers, either.  So we don't know anything about the Author's mathematical abilities. (But we can see that the Scribe did not know Euclidean geometry at all.)
He possessed some knowledge of an astro-medical practice — herbal, pharmaceutical, calendrical, balneological, and gynecological

My reference to his practical knowledge of using numbers was referred his potential process of estimating the content capacity for glyphs and encoding constraints. It definitely did not require mathematics but some good understanding how to use numbers. And also a potential exposure to ledger systems, which may have been a potential inspiration for the positional writing system, if ever.

(16-07-2026, 10:54 AM)Pointless.. Wrote: You are not allowed to view links. Register or Login to view.While they could have hired an outside illustrator, the final artwork would make that an impressive choice

If you compare the drawings of Herbal A and Herbal B, you can see differences in the style of the plant drawings, the coloring (which is sometimes more bold than necessary in Herbal B), and sometimes the size (which may be due to the scribe's small handwriting. He drew large plants to fill the page).
Apparently, the scribes were also artists.

(16-07-2026, 10:54 AM)Pointless.. Wrote: You are not allowed to view links. Register or Login to view.That said, I can certainly agree that the system as a whole does not match any known script

If desired, you can find similar glyphs of Latin letters in You are not allowed to view links. Register or Login to view.. This doesn't say much about the Voynichese symbols, but it does suggest that the author may have been familiar with Latin.
A special characteristic of VMS writing is that it is written quite quickly (if you write with a pen). All letters are written in a few strokes. Of course, you can say that Latin is similarly easy to write, but Voynichese still wins in comparison

(16-07-2026, 11:09 AM)ololololo Wrote: You are not allowed to view links. Register or Login to view.If you compare the drawings of Herbal A and Herbal B, you can see differences in the style of the plant drawings, the coloring (which is sometimes more bold than necessary in Herbal B), and sometimes the size (which may be due to the scribe's small handwriting. He drew large plants to fill the page).

I recall that this topic was discussed several months ago.  But I would like to object, your honor. Which is the proper thread?

All the best, --stolfi

(16-07-2026, 04:30 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.

(16-07-2026, 11:09 AM)ololololo Wrote: You are not allowed to view links. Register or Login to view.If you compare the drawings of Herbal A and Herbal B, you can see differences in the style of the plant drawings, the coloring (which is sometimes more bold than necessary in Herbal B), and sometimes the size (which may be due to the scribe's small handwriting. He drew large plants to fill the page).

I recall that this topic was discussed several months ago.  But I would like to object, your honor. Which is the proper thread?

All the best, --stolfi

The Voynich Manuscript's herbal section is traditionally divided into Herbal A and Herbal B, a distinction first proposed by Prescott Currier based on textual characteristics and scribal hands. [1][3][47] Currier noted a direct correlation between his "Language A" and Scribe 1, and "Language B" and Scribe 2, a finding later confirmed by Lisa Fagin Davis. [3][39] While the textual differences are well-established, researchers have also observed distinctions in the plant illustrations themselves. [1][15] On voynich.nu, René Zandbergen notes that consistent differences have been observed in the drawing style between Herbal A and Herbal B pages. [15] Sam G, on the voynich.ninja forum, highlighted the striking difference in the use of red coloring, which is abundant in Herbal A illustrations but largely absent in Herbal B, with You are not allowed to view links. Register or Login to view. being a possible exception due to later modification. [9][17] Koen Gheuens and Cary Rapaport, in a 2024 independent blog post, further explored whether plants could be sorted into Herbal A or B based on images alone, excluding color due to debates about its originality. [2][10] They found that in most cases, it is possible to assign a plant to Herbal A or Herbal B based on distinctive features of the drawings. [2] Nick Pelling, on Cipher Mysteries, observes that Herbal A pages appear to represent plants, while Herbal B pages, made later, seem to be something else entirely made to resemble a plant treatise. [12] However, D.N. O'Donovan, commenting on Gheuens' blog, expresses skepticism about reorganizing the manuscript based solely on the Currier A and B textual division, suggesting that a clearer understanding of the 'Currier divide' is needed. [5]

Sources:
[1] Special Topics: MS Illustrations — You are not allowed to view links. Register or Login to view.
[2] Can we sort plants in Herbal A and Herbal B based on images alone? — You are not allowed to view links. Register or Login to view.
[3] Three things Herbal A images do that Herbal B images don't — You are not allowed to view links. Register or Login to view.
[5] Can we sort plants in Herbal A and Herbal B based on images alone? — You are not allowed to view links. Register or Login to view.
[9] A and B plants — You are not allowed to view links. Register or Login to view.
[10] Can we sort plants in Herbal A and Herbal B based on images alone? — You are not allowed to view links. Register or Login to view.
[12] Q13 and Voynich balneology sources…? — You are not allowed to view links. Register or Login to view.
[15] Special Topics: Origin of the MS — You are not allowed to view links. Register or Login to view.
[17] Three things Herbal A images do that Herbal B images don't — You are not allowed to view links. Register or Login to view.
[39] Voynich Paleography — You are not allowed to view links. Register or Login to view.
[47] Captain Prescott H. Currier, Papers on the Voynich Manuscript (1976 Washington Seminar) — You are not allowed to view links. Register or Login to view.

(16-07-2026, 08:42 PM)DG97EEB Wrote: You are not allowed to view links. Register or Login to view.On voynich.nu, René Zandbergen notes that consistent differences have been observed in the drawing style between Herbal A and Herbal B pages.

I am just summarising the significant piece of work by Koen Gheuens and Cary Rapaport.
But I am more curious about the remark of (let me see: ) @ololololo:

(16-07-2026, 11:09 AM)ololololo Wrote: You are not allowed to view links. Register or Login to view.If you compare the drawings of Herbal A and Herbal B, you can see differences in the style of the plant drawings, the coloring (which is sometimes more bold than necessary in Herbal B), and sometimes the size (which may be due to the scribe's small handwriting. He drew large plants to fill the page).

Was that based on the same results of Koen and Cary, or an independent observation?

(16-07-2026, 04:30 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.

(16-07-2026, 11:09 AM)ololololo Wrote: You are not allowed to view links. Register or Login to view.If you compare the drawings of Herbal A and Herbal B, you can see differences in the style of the plant drawings, the coloring (which is sometimes more bold than necessary in Herbal B), and sometimes the size (which may be due to the scribe's small handwriting. He drew large plants to fill the page).

I recall that this topic was discussed several months ago.  But I would like to object, your honor. Which is the proper thread?

All the best, --stolfi

This discussion must have passed me by 
In general, I believe that scribes (either hired or authors) could draw and even color their illustrations. In my opinion, the style of these illustrations is very similar between A and B.
It is possible that scribe B did not have a good imagination when it came to creating plants, as he copied his own illustrations.
For example, sunflowers: f33v-f40v-f50r; one-sided plants: f31r-f31v-f46r; also f39v-f40r and f31r-f43r (look at flowers).

---

(17-07-2026, 12:24 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.Was that based on the same results of Koen and Cary, or an independent observation?

I don't know how independent it is, but I came to these conclusions on my own

The difference I noticed is in the design of the drawings:
[attachment=16646]
[attachment=16647]
The green paint seems to have been applied carefully, and it looks a little dry. Despite this, the drawing is quite clear.
[attachment=16648]
[attachment=16649]
From these drawings, we can see that they are generally poorly executed. The paint is either too dull or too thickly applied (as seen on the leaves of the second plant).
Differences in the shape of the plants? In Herbal B, they often look larger, which I think matches the scribe's very small handwriting. I talked about this inYou are not allowed to view links. Register or Login to view..

The reason I asked is because, until Koen and Cary produced their results, really no significant trend had been observed by anyone, over the preceding decade. The only known point was (IIRC) a single line cutting across plant stems, and one may be sceptical whether such a detail is significant.

Anyway, now my question is: do these two points hold consistently between plants on A and B pages?