<![CDATA[The Voynich Ninja - Analysis of the text]]> https://www.voynich.ninja/ Wed, 29 Jul 2026 09:50:32 +0000 MyBB <![CDATA[VMS stats: YT video]]> https://www.voynich.ninja/thread-5974.html Tue, 28 Jul 2026 06:00:23 +0000 elemental]]> https://www.voynich.ninja/thread-5974.html  Enjoy. 

]]>  Enjoy. 

]]> <![CDATA[Paragraph layout]]> https://www.voynich.ninja/thread-5970.html Mon, 27 Jul 2026 20:20:04 +0000 davidjackson]]> https://www.voynich.ninja/thread-5970.html If we look at standard manuscripts or incunabula (usually agreed to be visually the same as their hand written equivalents) from, say, 1450 - 1500 we see text written in blocks, not paragraphs. 
Many manuscripts will have clear paragraph markers -pilcrows-, usually distinguishable by a different colour or size, to break up information units, rather than the modern usage of using a line break, as I am doing in this post. You are not allowed to view links. Register or Login to view..

The Voynich, from the very first page, uses the line break format accompanied by a pilcrow.
[Image: image.jpg?ref=f1r&q=f1r-256-202-1027-123...&caption=1]

This habit is clear throughout the manuscript and all scribes follow the same layout instruction
[Image: image.jpg?ref=f43v&q=f43v-417-164-982-60...&caption=1]

It is striking that the same layout instructions were applied by all scribes across all sections of the manuscript. A template, even if only verbal instructions, was given for layout and this was applied across the manuscript. The same layout was used for page after page, this indicates a conscience decision and even a group instruction issued if indeed multiple scribes were used
[Image: image.jpg?ref=f115r&q=f115r-101-66-1181-...&caption=1]

How did paragraphs work?

Quote:But while distinction of words was disregarded in early literary texts, distinction of important pauses in the sense was recognized from the first. The papyrus of the Persae of Timotheus of Miletus, the oldest MS. of a Greek classic in existence, of the end of the 4th century B.C., is written in independent paragraphs. This is a natural system, the simplicity of which has caused it to be the system of modern times. But, in addition, the Greek scribe also separated paragraphs by inserting a short horizontal stroke, παράγραφος, between them at the commencement of the lines of writing. It should be noted that this stroke indicated the close of a passage, and therefore belonged to the paragraph just concluded, and did not stand for an initial sign for the new paragraph which followed. The dividing stroke was also used to mark off the different speeches of a play. Besides the stroke, a wedge-shaped sign or tick might be used. But to make every paragraph stand distinctly by itself would have entailed a certain loss of space. If the concluding line were short, there would remain a long space unfilled. Therefore, when this occurred, it became customary to leave only a short space blank to mark the termination of the paragraph, and then to proceed with the new paragraph in the same line, the παράγραφος at the same time preventing possible ambiguity. The next step was to project the first letter of the first full line of the new paragraph slightly into the margin, as a still further distinction; and lastly to enlarge it. The enlargement of the letter gave it so much prominence that the dividing stroke could then be dispensed with, and in this form the new paragraph was henceforward indicated in Greek MSS., it being immaterial whether the enlarged letter was the initial or a medial letter of a word. As early as the 5th century there is evidence that the παράγραφος was losing its meaning with the scribes, for in the Codex Alexandrinus of the Bible it is not infrequently found in anomalous positions, particularly above the initial letters of different books, as if it were a mere ornament.
Encyclopedia Britannica - Manuscript entry 

Quote:The paragraph mark, enlarged initial, and disposition of text on the page formed part of an integrated system for indicating textual articulation. Medieval scribes indicated divisions in a text by combining graphic devices such as enlarged initials, paragraph signs, and the arrangement of lines on the page. These visual cues enabled readers to recognise the structure of the text without relying solely on punctuation. Malcolm B. Parkes — Pause and Effect p.65

Quote:Colometry. A modification of the practice of writing the text continuously was allowed in the case of certain works. Rhetorical texts, such as the orations of Demosthenes and Cicero, and the text of the Bible, might be broken up into short clauses or sense-lines, apparently with the view of assisting reading aloud. Instances of MSS. so written are still extant. This system, to which the name of “colometry” has been given, is the arrangement by cola and commata referred to by St Jerome in his preface to Isaiah. It will be found more fully explained under the heading of Stichometry; where also is described the mechanical computation of the length of a text by measured lines, for the purpose or calculating the pay of the scribe. Encyclopedia Britannica - Manuscript entry

Medieval scribes were extremely reluctant to waste expensive parchment, so paragraph identification was almost always redundant. They combined several cues:
  • new line,
  • enlarged initial,
  • rubrication,
  • paragraph mark,
  • marginal projection.

A new line by itself was generally considered "too subtle" unless for the reason of Colometry.

The Voynich often signals a paragraph by:
  • beginning on a fresh line,
  • using an enlarged "gallows" glyph,
  • leaving a visibly short final line in the previous paragraph without bothering about justification or waste of parchment.

Notably, it does not employ coloured initials or rubricated pilcrows, which were among the commonest medieval paragraph markers and which would have been instantly identifiable to any medieval reader. That makes its paragraphing visually simpler than many contemporary manuscripts while still conforming to the broader medieval practice of marking new textual units through line breaks and paragraph-initial emphasis.

What am I getting at?

If we discard the idea that the Voynich was designed to be read aloud - clearly most of it was not - then what influence caused the authors to choose and stay with this particular layout throughout the whole of the book?

I think the closest structural comparators are not illuminated or library books but plain working manuscripts—medical recipe books, university notebooks, and practical handbooks copied between about 1400 and 1450. We often see new-line paragraphing with minimal decoration and no attempt at justification.

What style could the scribes have been influenced by?
  • Medical recipe books (Receptaria) 
  • Astrological almanacs with prognostications 
  • Practical herbals – especially those copied by working physicians rather than produced as luxury manuscripts

These genres routinely consist of dozens or hundreds of short, self-contained entries beginning on a new line, and they are among the closest surviving medieval layouts to the Voynich's "recipe" section. We also see many clues to the Voynich being influenced by these working books, for example, many years ago we chased down examples of the Zodiac being influenced by the German hausbach calendar tradition, especially the crossbowman. This forum contains many examples.

In short, the layout and visual clues in the Voynich points to the authors being influenced more by working books than by library books. 

However, in many ways the Voynich has a much higher "production value" than these working books. It is much longer than most working books, many of which would have been produced by just one scribe. 

The images are of a higher quality. The layout of the text is much clearer and more professional. There appear to have been a team of production scribes working on the book. The quality of the parchment is higher and there is no real attempt to maximise the use of the parchment real estate. Multiple colour inks are used with care and attention to detail.

We are, once again, faced with a paradox. The Voynich layout and style is influenced by working books; its use of materials and multiple scribes suggests high end resources were available for its production.]]> If we look at standard manuscripts or incunabula (usually agreed to be visually the same as their hand written equivalents) from, say, 1450 - 1500 we see text written in blocks, not paragraphs. 
Many manuscripts will have clear paragraph markers -pilcrows-, usually distinguishable by a different colour or size, to break up information units, rather than the modern usage of using a line break, as I am doing in this post. You are not allowed to view links. Register or Login to view..

The Voynich, from the very first page, uses the line break format accompanied by a pilcrow.
[Image: image.jpg?ref=f1r&q=f1r-256-202-1027-123...&caption=1]

This habit is clear throughout the manuscript and all scribes follow the same layout instruction
[Image: image.jpg?ref=f43v&q=f43v-417-164-982-60...&caption=1]

It is striking that the same layout instructions were applied by all scribes across all sections of the manuscript. A template, even if only verbal instructions, was given for layout and this was applied across the manuscript. The same layout was used for page after page, this indicates a conscience decision and even a group instruction issued if indeed multiple scribes were used
[Image: image.jpg?ref=f115r&q=f115r-101-66-1181-...&caption=1]

How did paragraphs work?

Quote:But while distinction of words was disregarded in early literary texts, distinction of important pauses in the sense was recognized from the first. The papyrus of the Persae of Timotheus of Miletus, the oldest MS. of a Greek classic in existence, of the end of the 4th century B.C., is written in independent paragraphs. This is a natural system, the simplicity of which has caused it to be the system of modern times. But, in addition, the Greek scribe also separated paragraphs by inserting a short horizontal stroke, παράγραφος, between them at the commencement of the lines of writing. It should be noted that this stroke indicated the close of a passage, and therefore belonged to the paragraph just concluded, and did not stand for an initial sign for the new paragraph which followed. The dividing stroke was also used to mark off the different speeches of a play. Besides the stroke, a wedge-shaped sign or tick might be used. But to make every paragraph stand distinctly by itself would have entailed a certain loss of space. If the concluding line were short, there would remain a long space unfilled. Therefore, when this occurred, it became customary to leave only a short space blank to mark the termination of the paragraph, and then to proceed with the new paragraph in the same line, the παράγραφος at the same time preventing possible ambiguity. The next step was to project the first letter of the first full line of the new paragraph slightly into the margin, as a still further distinction; and lastly to enlarge it. The enlargement of the letter gave it so much prominence that the dividing stroke could then be dispensed with, and in this form the new paragraph was henceforward indicated in Greek MSS., it being immaterial whether the enlarged letter was the initial or a medial letter of a word. As early as the 5th century there is evidence that the παράγραφος was losing its meaning with the scribes, for in the Codex Alexandrinus of the Bible it is not infrequently found in anomalous positions, particularly above the initial letters of different books, as if it were a mere ornament.
Encyclopedia Britannica - Manuscript entry 

Quote:The paragraph mark, enlarged initial, and disposition of text on the page formed part of an integrated system for indicating textual articulation. Medieval scribes indicated divisions in a text by combining graphic devices such as enlarged initials, paragraph signs, and the arrangement of lines on the page. These visual cues enabled readers to recognise the structure of the text without relying solely on punctuation. Malcolm B. Parkes — Pause and Effect p.65

Quote:Colometry. A modification of the practice of writing the text continuously was allowed in the case of certain works. Rhetorical texts, such as the orations of Demosthenes and Cicero, and the text of the Bible, might be broken up into short clauses or sense-lines, apparently with the view of assisting reading aloud. Instances of MSS. so written are still extant. This system, to which the name of “colometry” has been given, is the arrangement by cola and commata referred to by St Jerome in his preface to Isaiah. It will be found more fully explained under the heading of Stichometry; where also is described the mechanical computation of the length of a text by measured lines, for the purpose or calculating the pay of the scribe. Encyclopedia Britannica - Manuscript entry

Medieval scribes were extremely reluctant to waste expensive parchment, so paragraph identification was almost always redundant. They combined several cues:
  • new line,
  • enlarged initial,
  • rubrication,
  • paragraph mark,
  • marginal projection.

A new line by itself was generally considered "too subtle" unless for the reason of Colometry.

The Voynich often signals a paragraph by:
  • beginning on a fresh line,
  • using an enlarged "gallows" glyph,
  • leaving a visibly short final line in the previous paragraph without bothering about justification or waste of parchment.

Notably, it does not employ coloured initials or rubricated pilcrows, which were among the commonest medieval paragraph markers and which would have been instantly identifiable to any medieval reader. That makes its paragraphing visually simpler than many contemporary manuscripts while still conforming to the broader medieval practice of marking new textual units through line breaks and paragraph-initial emphasis.

What am I getting at?

If we discard the idea that the Voynich was designed to be read aloud - clearly most of it was not - then what influence caused the authors to choose and stay with this particular layout throughout the whole of the book?

I think the closest structural comparators are not illuminated or library books but plain working manuscripts—medical recipe books, university notebooks, and practical handbooks copied between about 1400 and 1450. We often see new-line paragraphing with minimal decoration and no attempt at justification.

What style could the scribes have been influenced by?
  • Medical recipe books (Receptaria) 
  • Astrological almanacs with prognostications 
  • Practical herbals – especially those copied by working physicians rather than produced as luxury manuscripts

These genres routinely consist of dozens or hundreds of short, self-contained entries beginning on a new line, and they are among the closest surviving medieval layouts to the Voynich's "recipe" section. We also see many clues to the Voynich being influenced by these working books, for example, many years ago we chased down examples of the Zodiac being influenced by the German hausbach calendar tradition, especially the crossbowman. This forum contains many examples.

In short, the layout and visual clues in the Voynich points to the authors being influenced more by working books than by library books. 

However, in many ways the Voynich has a much higher "production value" than these working books. It is much longer than most working books, many of which would have been produced by just one scribe. 

The images are of a higher quality. The layout of the text is much clearer and more professional. There appear to have been a team of production scribes working on the book. The quality of the parchment is higher and there is no real attempt to maximise the use of the parchment real estate. Multiple colour inks are used with care and attention to detail.

We are, once again, faced with a paradox. The Voynich layout and style is influenced by working books; its use of materials and multiple scribes suggests high end resources were available for its production.]]> <![CDATA[Word boundaries "dl"]]> https://www.voynich.ninja/thread-5957.html Sat, 25 Jul 2026 12:03:51 +0000 ololololo]]> https://www.voynich.ninja/thread-5957.html Register or Login to view., I noticed a peculiar feature when comparing words with You are not allowed to view links. Register or Login to view. with the text of f104r.28: sometimes the boundary between words was the combination of dl. Of course, my work is still somewhat questionable, but I found this observation interesting.
I decided to test the behavior of this combination in a manuscript. It appears in both Currier A and Currier B. In both cases, the combinations behave in the same way - either they stand alone or they are combined with words like ch (e.g., chdlety f47v).
In the recipe section, this combination is often found next to d. Examples: daindl, dlchd, opcheddl, ckhddl, oteeodl.
Before dl, there may be an o, and in one case they form one word together (odl, also often in the recipe section), while in the other case the previous word ends with o (examples: tocheo dlchy; oteeodl)
Of course, these observations are not enough to draw any conclusions, but we can make some assumptions.
It is possible that the combination of dl carries the meaning of distinguishing between words. It is also possible that this combination arises from the division of text into combinations of letters (as in many classical ciphers, the text is first divided into bigrams or trigrams).

What do you think?]]> Register or Login to view., I noticed a peculiar feature when comparing words with You are not allowed to view links. Register or Login to view. with the text of f104r.28: sometimes the boundary between words was the combination of dl. Of course, my work is still somewhat questionable, but I found this observation interesting.
I decided to test the behavior of this combination in a manuscript. It appears in both Currier A and Currier B. In both cases, the combinations behave in the same way - either they stand alone or they are combined with words like ch (e.g., chdlety f47v).
In the recipe section, this combination is often found next to d. Examples: daindl, dlchd, opcheddl, ckhddl, oteeodl.
Before dl, there may be an o, and in one case they form one word together (odl, also often in the recipe section), while in the other case the previous word ends with o (examples: tocheo dlchy; oteeodl)
Of course, these observations are not enough to draw any conclusions, but we can make some assumptions.
It is possible that the combination of dl carries the meaning of distinguishing between words. It is also possible that this combination arises from the division of text into combinations of letters (as in many classical ciphers, the text is first divided into bigrams or trigrams).

What do you think?]]> <![CDATA[The o's in VMS]]> https://www.voynich.ninja/thread-5947.html Wed, 22 Jul 2026 01:11:57 +0000 bvanheu]]> https://www.voynich.ninja/thread-5947.html
I can't help but notice there are different kind of ways O's are drawn in VMS. I always starts my O from the top going counterclockwise. I've exagerated the traits in KolourPaint to show the starting point and direction. There doesn't seem to be a pattern (e.g. whole page written with O from bottom). The most common O seems to be starting from the top.

Starting from the top


.png   top_o_1.png (Size: 28.79 KB / Downloads: 202)

.png   top_o_2.png (Size: 19.39 KB / Downloads: 312)

.png   top_o_3.png (Size: 78.94 KB / Downloads: 202)

Starting from bottom


.png   bottom_o_1.png (Size: 16.52 KB / Downloads: 311)

.png   bottom_o_2.png (Size: 62.33 KB / Downloads: 201)

.png   bottom_o_3.png (Size: 38.59 KB / Downloads: 202)

.png   bottom_o_4.png (Size: 25.67 KB / Downloads: 313)

In two quill traits

These could be EVA a though?


.png   weird_o_1.png (Size: 26.28 KB / Downloads: 311)

.png   weird_o_2.png (Size: 37.41 KB / Downloads: 202)

.png   weird_o_3.png (Size: 21.59 KB / Downloads: 313)

.png   weird_o_4.png (Size: 25.55 KB / Downloads: 312) ]]>
I can't help but notice there are different kind of ways O's are drawn in VMS. I always starts my O from the top going counterclockwise. I've exagerated the traits in KolourPaint to show the starting point and direction. There doesn't seem to be a pattern (e.g. whole page written with O from bottom). The most common O seems to be starting from the top.

Starting from the top


.png   top_o_1.png (Size: 28.79 KB / Downloads: 202)

.png   top_o_2.png (Size: 19.39 KB / Downloads: 312)

.png   top_o_3.png (Size: 78.94 KB / Downloads: 202)

Starting from bottom


.png   bottom_o_1.png (Size: 16.52 KB / Downloads: 311)

.png   bottom_o_2.png (Size: 62.33 KB / Downloads: 201)

.png   bottom_o_3.png (Size: 38.59 KB / Downloads: 202)

.png   bottom_o_4.png (Size: 25.67 KB / Downloads: 313)

In two quill traits

These could be EVA a though?


.png   weird_o_1.png (Size: 26.28 KB / Downloads: 311)

.png   weird_o_2.png (Size: 37.41 KB / Downloads: 202)

.png   weird_o_3.png (Size: 21.59 KB / Downloads: 313)

.png   weird_o_4.png (Size: 25.55 KB / Downloads: 312) ]]> <![CDATA[Non-linear line ordering]]> https://www.voynich.ninja/thread-5943.html Sun, 19 Jul 2026 15:41:08 +0000 bvanheu]]> https://www.voynich.ninja/thread-5943.html Register or Login to view. and f58v, there seem to be evidence that some lines were written after the immediate bottom line, with glyph from the top line being written over the glyph from the bottom line?

If the gallow was written after the top line, why make it so big that it extends over the above glyph?


.png   f58v_2.png (Size: 136.83 KB / Downloads: 185)

Why start the gallow line so that it's *over* the glyph above?


.png   20260719_11h20m51s_grim.png (Size: 105.36 KB / Downloads: 184)

It seems to me the tail has been made in one continuous trait?


.png   20260719_11h19m26s_grim.png (Size: 113.14 KB / Downloads: 184)

Is there a concensus around line ordering, beside that "block of text" were written before/after others?

.png   20260719_11h35m25s_grim.png (Size: 121.22 KB / Downloads: 171) ]]> Register or Login to view. and f58v, there seem to be evidence that some lines were written after the immediate bottom line, with glyph from the top line being written over the glyph from the bottom line?

If the gallow was written after the top line, why make it so big that it extends over the above glyph?


.png   f58v_2.png (Size: 136.83 KB / Downloads: 185)

Why start the gallow line so that it's *over* the glyph above?


.png   20260719_11h20m51s_grim.png (Size: 105.36 KB / Downloads: 184)

It seems to me the tail has been made in one continuous trait?


.png   20260719_11h19m26s_grim.png (Size: 113.14 KB / Downloads: 184)

Is there a concensus around line ordering, beside that "block of text" were written before/after others?

.png   20260719_11h35m25s_grim.png (Size: 121.22 KB / Downloads: 171) ]]> <![CDATA[m on side for Michał f42v, W on side for Wojnicz 1r]]> https://www.voynich.ninja/thread-5926.html Thu, 16 Jul 2026 04:08:46 +0000 conlangyalesbaby]]> https://www.voynich.ninja/thread-5926.html .jpg   M for Michal.JPG (Size: 6.79 KB / Downloads: 248)

The sh glyph has an m on its side You are not allowed to view links. Register or Login to view..  Is the evidence for Michał Habdank-Wojnicz?


.jpg   W for x.JPG (Size: 25.17 KB / Downloads: 247)

Also for You are not allowed to view links. Register or Login to view. there is a w on it's side for Pi upside down. Could it be Wojnicz?  That would be the x glyph!]]> .jpg   M for Michal.JPG (Size: 6.79 KB / Downloads: 248)

The sh glyph has an m on its side You are not allowed to view links. Register or Login to view..  Is the evidence for Michał Habdank-Wojnicz?


.jpg   W for x.JPG (Size: 25.17 KB / Downloads: 247)

Also for You are not allowed to view links. Register or Login to view. there is a w on it's side for Pi upside down. Could it be Wojnicz?  That would be the x glyph!]]> <![CDATA[[split] Designing the glyph system]]> https://www.voynich.ninja/thread-5929.html Wed, 15 Jul 2026 07:32:36 +0000 Pointless..]]> https://www.voynich.ninja/thread-5929.html
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")]]>
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")]]> <![CDATA[Least effort generator]]> https://www.voynich.ninja/thread-5920.html Mon, 13 Jul 2026 20:21:57 +0000 tikonen]]> https://www.voynich.ninja/thread-5920.html
My working assumption is that the manuscript is a forgery made to part rich people of their money.  I've been thinking how it was generated in seemingly consistent way. Maybe it was done by sweat shopping few cleric students for pocket money and minimal time and and money was invested to train them.

So I guess there was possibly a template or a collection of tokens the scribes used to learn to generate VM's fantasy words with some personal twist, mistakes and tweaks. The binomial distribution of a word length gives some support to this idea, indicating that the word lengths may be random. After some practice the tokens were not needed anymore and scribe would just make it on the fly. VM's high word "kindness" to the previous rows could also support this idea.

To test the feasibility of this idea I wrote a simple program that when given a word from VM it finds a minimal number of predefined tokens in order to construct the word. I did not consider words that are used only twice or less and the ones with rare letters. 

Here is and example of one token list set I conjured.  
Word is constructed by picking a token from each list (or skipping it) and concatenating them to a word. 
As you can guess list s0 is considered first when building a word, then list s1 and so on until s5 as last. A token from each list can be used only once for a word. 
This limited example already covers (with above mentioned filtering) 34% of all the unique words and 75% of all words (many words are used multiple times). With this set the average number of needed tokens for common words with >=50 instances is 2.1. (For >=10 it's 2.5). So basically one can generate most VM words by picking just few tokens.

s0 = ['qok', 'qoke', 'qot', 'ot', 'd', 'q', 'l', 'yk']
s1 = ['ch', 'sh', 's', 'ok']
s2 = ['e', 'o', 'r', 'k']
s3 = ['r', 'k', 'o', 'e']
s4 = ['cth', 'ckh']
s5 = ['daiin', 'aiin', 'ain', 'ey', 'edy', 'dy', 'y', 'ar', 'or', 'al', 'am', 'air', 'ody', 'ol']

Example: Word 'chedy' would be token 'ch' from list s1 and 'edy' from s5.

Position on each token in each list reflects how often it's used in VM words. Each word may have multiple "solutions", algorithm picks shortest.  
It's relatively easy to make a larger set for >95% word coverage but then of course the average number of token picks increases (slowly). 

Some points:
- One can generate non VM words if tokens are chosen totally randomly. Some external rules are required. 
- Real VM words do not use much s2 and s3. Omitting them drops coverage from 75% to 66%.
- Impossible to draw line how much is template and how much is just random on the fly variation of scribes
- There are some words that have high occurrence but complicate the set, they need a token that is used mostly on that word. 
 
I don't know if this has any merit but at least it's an interesting exercise.]]>
My working assumption is that the manuscript is a forgery made to part rich people of their money.  I've been thinking how it was generated in seemingly consistent way. Maybe it was done by sweat shopping few cleric students for pocket money and minimal time and and money was invested to train them.

So I guess there was possibly a template or a collection of tokens the scribes used to learn to generate VM's fantasy words with some personal twist, mistakes and tweaks. The binomial distribution of a word length gives some support to this idea, indicating that the word lengths may be random. After some practice the tokens were not needed anymore and scribe would just make it on the fly. VM's high word "kindness" to the previous rows could also support this idea.

To test the feasibility of this idea I wrote a simple program that when given a word from VM it finds a minimal number of predefined tokens in order to construct the word. I did not consider words that are used only twice or less and the ones with rare letters. 

Here is and example of one token list set I conjured.  
Word is constructed by picking a token from each list (or skipping it) and concatenating them to a word. 
As you can guess list s0 is considered first when building a word, then list s1 and so on until s5 as last. A token from each list can be used only once for a word. 
This limited example already covers (with above mentioned filtering) 34% of all the unique words and 75% of all words (many words are used multiple times). With this set the average number of needed tokens for common words with >=50 instances is 2.1. (For >=10 it's 2.5). So basically one can generate most VM words by picking just few tokens.

s0 = ['qok', 'qoke', 'qot', 'ot', 'd', 'q', 'l', 'yk']
s1 = ['ch', 'sh', 's', 'ok']
s2 = ['e', 'o', 'r', 'k']
s3 = ['r', 'k', 'o', 'e']
s4 = ['cth', 'ckh']
s5 = ['daiin', 'aiin', 'ain', 'ey', 'edy', 'dy', 'y', 'ar', 'or', 'al', 'am', 'air', 'ody', 'ol']

Example: Word 'chedy' would be token 'ch' from list s1 and 'edy' from s5.

Position on each token in each list reflects how often it's used in VM words. Each word may have multiple "solutions", algorithm picks shortest.  
It's relatively easy to make a larger set for >95% word coverage but then of course the average number of token picks increases (slowly). 

Some points:
- One can generate non VM words if tokens are chosen totally randomly. Some external rules are required. 
- Real VM words do not use much s2 and s3. Omitting them drops coverage from 75% to 66%.
- Impossible to draw line how much is template and how much is just random on the fly variation of scribes
- There are some words that have high occurrence but complicate the set, they need a token that is used mostly on that word. 
 
I don't know if this has any merit but at least it's an interesting exercise.]]> <![CDATA[EVA "qo" in 15th century Spanish cipher]]> https://www.voynich.ninja/thread-5895.html Fri, 10 Jul 2026 12:46:52 +0000 JustAnotherTheory]]> https://www.voynich.ninja/thread-5895.html

.png   321654.png (Size: 11.44 KB / Downloads: 182)

The documents are the following:
  • Barcelona,Archivo de la Corona de Aragón, Diversos, Sástago, 193 (Lío B), nº 79/58.
  • Barcelona,Archivo de la Corona de Aragón, Diversos, Sástago, 193 (Lío B), nº 79/59.

Unfortunately, the rest of the documents do not seem at first glance to bear ressemblance to the VMS. Here is a random example from one of them:


.png   987543.png (Size: 567.55 KB / Downloads: 185)

________________________________________

Links to the docs:

You are not allowed to view links. Register or Login to view.

You are not allowed to view links. Register or Login to view.]]>

.png   321654.png (Size: 11.44 KB / Downloads: 182)

The documents are the following:
  • Barcelona,Archivo de la Corona de Aragón, Diversos, Sástago, 193 (Lío B), nº 79/58.
  • Barcelona,Archivo de la Corona de Aragón, Diversos, Sástago, 193 (Lío B), nº 79/59.

Unfortunately, the rest of the documents do not seem at first glance to bear ressemblance to the VMS. Here is a random example from one of them:


.png   987543.png (Size: 567.55 KB / Downloads: 185)

________________________________________

Links to the docs:

You are not allowed to view links. Register or Login to view.

You are not allowed to view links. Register or Login to view.]]> <![CDATA[Split: Entropy digression]]> https://www.voynich.ninja/thread-5894.html Wed, 08 Jul 2026 00:08:17 +0000 ReneZ]]> https://www.voynich.ninja/thread-5894.html (07-07-2026, 09:39 PM)Stefan Wirtz_2 Wrote: You are not allowed to view links. Register or Login to view.Those entropy calculations are still very depending upon the type of transliteration;
some more aggressive "experts" here are favourizing an "h2" of 2.3 for Voynichese, while the change of transliteration file can produce a value of ~2.9.
As @nablator already pointed out, this is not correct. You wrote that this value was reported for the v101 transliteration. I just recomputed this with my own tools, stripping the transliteration of all annotations, but keeping the full character set.

The v101 file uses 158 different characters, if one counts the space also as a character.
 
Computing the entropy values, also counting space as a character yields:
Single char entropy     H1 =  4.036
Character pair entropy H2 =  6.608
Conditional ch entropy h2 =   2.572

More realistic is not to count space as a character (it isn't). Then:
Single char entropy     H1 =  4.131
Character pair entropy H2 =  6.501
Conditional ch entropy h2 =   2.370

In this case, character pairs spanning a space are not included in the statistics.

The most representative value is 2.37.

Comparison values may be found here: You are not allowed to view links. Register or Login to view.
and here: You are not allowed to view links. Register or Login to view.]]> (07-07-2026, 09:39 PM)Stefan Wirtz_2 Wrote: You are not allowed to view links. Register or Login to view.Those entropy calculations are still very depending upon the type of transliteration;
some more aggressive "experts" here are favourizing an "h2" of 2.3 for Voynichese, while the change of transliteration file can produce a value of ~2.9.
As @nablator already pointed out, this is not correct. You wrote that this value was reported for the v101 transliteration. I just recomputed this with my own tools, stripping the transliteration of all annotations, but keeping the full character set.

The v101 file uses 158 different characters, if one counts the space also as a character.
 
Computing the entropy values, also counting space as a character yields:
Single char entropy     H1 =  4.036
Character pair entropy H2 =  6.608
Conditional ch entropy h2 =   2.572

More realistic is not to count space as a character (it isn't). Then:
Single char entropy     H1 =  4.131
Character pair entropy H2 =  6.501
Conditional ch entropy h2 =   2.370

In this case, character pairs spanning a space are not included in the statistics.

The most representative value is 2.37.

Comparison values may be found here: You are not allowed to view links. Register or Login to view.
and here: You are not allowed to view links. Register or Login to view.]]> <![CDATA[The fifth gallow?]]> https://www.voynich.ninja/thread-5887.html Tue, 07 Jul 2026 16:10:48 +0000 ololololo]]> https://www.voynich.ninja/thread-5887.html
.jpg   36.jpg (Size: 12.74 KB / Downloads: 153)

.jpg   47.jpg (Size: 26.03 KB / Downloads: 153)

.jpg   3.jpg (Size: 24.48 KB / Downloads: 152)


There were also his appearances in botany, but I couldn't find them. Most often, there is no additional loop (as in the first two screenshots). 
Outwardly, it looks vaguely like an f with a second leg, but I think that's not the case, because we already have an f with a second leg:

.jpg   Снимок экрана 2026-07-07 191018.jpg (Size: 27.6 KB / Downloads: 152)

What do you think?]]>
.jpg   36.jpg (Size: 12.74 KB / Downloads: 153)

.jpg   47.jpg (Size: 26.03 KB / Downloads: 153)

.jpg   3.jpg (Size: 24.48 KB / Downloads: 152)


There were also his appearances in botany, but I couldn't find them. Most often, there is no additional loop (as in the first two screenshots). 
Outwardly, it looks vaguely like an f with a second leg, but I think that's not the case, because we already have an f with a second leg:

.jpg   Снимок экрана 2026-07-07 191018.jpg (Size: 27.6 KB / Downloads: 152)

What do you think?]]> <![CDATA[Statistical analysis, two findings, one potentially new]]> https://www.voynich.ninja/thread-5878.html Sun, 05 Jul 2026 00:50:44 +0000 mjtemkin]]> https://www.voynich.ninja/thread-5878.html
All of the work is publicly available, including source code, a pdf writeup, and a step by step guide to what I did.

The work is available here: You are not allowed to view links. Register or Login to view.
The pdf is here and attached to this post: You are not allowed to view links. Register or Login to view.

What this is and isn't: this is not a decipherment, nor was an attempt to decipher the manuscript. It's an independent replication and methodological extension of existing statistical work asking the question "was this manuscript generated from some kind of natural language." All code is public and every result comes with a permutation-based null model, so everything is reproducible from scratch.

Two main findings:

1. Immediate word-doubling at ~2x chance (word[i] == word[i+1], within a manuscript line). This is a well-known anomaly, but I tested it more carefully than I'd seen done before, across six languages and registers with formal z-scores throughout. I used some language samples that I thought could potentially contain word doublings naturally, to compare against the manuscript, including:

  - Culpeper's Complete Herbal (English, genre-matched): zero doublings, far below its own chance baseline. The hypothesis here was that if the manuscript was some kind of list (recipes, etc) perhaps a contemporaneous similar list could show the same pattern.
  - Carmina Burana In Taberna (Latin verse, maximally repetition-saturated, "bibit" repeated 24 times consecutively): zero doublings. The hypothesis here was that perhaps a contemporaneous document that was more poetic in nature could show similar doublings.
  - Arabic Al-Baqarah and Yusuf (consonantal): at or below chance. Arabic and Hebrew were including to look at languages that don't contain consonants and could show patterns that don't exist in long-form Latin languages because words could have multiple meanings.
  - Hebrew Psalms 113-150 / Hallel (liturgical Hebrew poetry): z = -2.87, p = 0.998- significantly SUPPRESSED below chance. The hypothesis here was that perhaps poetry in a language with a different structure could show similar doublings.
  - Hebrew Psalms full 150 chapters (19,662 words): z = -3.73, p = 1.000- even more strongly suppressed
  - Sanskrit Rigveda (complete, 135,279 words, IAST romanization): z = +3.04, p = 0.002- elevated above chance, though we flag this needs verification due to potential IAST sandhi encoding artifacts. Same hypothesis as liturgical Hebrew.

The Hebrew liturgical result is the one I find most interesting. Psalm 136 repeats the same phrase 26 times. Psalm 113 opens "הללו יה הללו". And yet the doubling rate is significantly BELOW chance — because real liturgical repetition always inserts at least one different word between repetitions of
the same phrase. "הללו יה הללו" is not "הללו הללו." That's also exactly what Carmina Burana does: "bibit hera, bibit herus-" same verb, different subject every time. Real repetition, whether poetic or liturgical, doesn't produce literal self-adjacency. Voynichese does, at ~2x chance, stationarily across the full manuscript.

The effect also survived a cross-section permutation test (p = 0.144 for the section gap — consistent with a single stationary process throughout), and a
local scramble test showing it depends on line co-membership rather than exact write-order.

2. Word-class ordering asymmetry — new to our analysis. Classifying words by their 2-character suffix (following Stolfi's grammar framework) and measuring directional ordering preferences between classes, both Currier languages show significantly asymmetric ordering: z = 4.72 for Language A, z = 4.77 for Language B, both p < 0.0001, robust to a frequency-matching control. Every other structural test returned null — this is the first positive,
grammar-suggestive result. I want to be careful: positional structure doesn't prove meaningful language. But it is something simple stateless generators don't produce by default, and the effect is equally strong in both Currier languages on the full validated corpus.

On the corpus: I used the complete FSG transliteration (Reeds 1994, from voynich.nu), validated word-for-word against the source file. In the process
we found that several Stars-section folios in our working corpus had been severely truncated (some folios had only 20-25% of their true content).
After rebuilding from the full authoritative file the corpus grew from ~14,000 to ~33,600 words (~88% of the manuscript). All findings were re-validated on the full corpus; the word-class asymmetry finding in particular changed meaningfully — what had looked like a gap between Language A and Language B disappeared entirely, with both now showing indistinguishable z-scores.

Happy to discuss methodology, findings, or criticisms. Hopefully it's clear what does and doesn't hold up (the Sanskrit result needs verification; an
earlier attempt to add corpus data produced fabricated text that was caught and removed before any analysis used it, and is documented in the repo). The GitHub repo has full documentation of every finding including the ones that failed or changed.

Thanks everyone, and looking forward to discussing.

.pdf   voynich_statistical_analysis.pdf (Size: 470.33 KB / Downloads: 16) ]]>
All of the work is publicly available, including source code, a pdf writeup, and a step by step guide to what I did.

The work is available here: You are not allowed to view links. Register or Login to view.
The pdf is here and attached to this post: You are not allowed to view links. Register or Login to view.

What this is and isn't: this is not a decipherment, nor was an attempt to decipher the manuscript. It's an independent replication and methodological extension of existing statistical work asking the question "was this manuscript generated from some kind of natural language." All code is public and every result comes with a permutation-based null model, so everything is reproducible from scratch.

Two main findings:

1. Immediate word-doubling at ~2x chance (word[i] == word[i+1], within a manuscript line). This is a well-known anomaly, but I tested it more carefully than I'd seen done before, across six languages and registers with formal z-scores throughout. I used some language samples that I thought could potentially contain word doublings naturally, to compare against the manuscript, including:

  - Culpeper's Complete Herbal (English, genre-matched): zero doublings, far below its own chance baseline. The hypothesis here was that if the manuscript was some kind of list (recipes, etc) perhaps a contemporaneous similar list could show the same pattern.
  - Carmina Burana In Taberna (Latin verse, maximally repetition-saturated, "bibit" repeated 24 times consecutively): zero doublings. The hypothesis here was that perhaps a contemporaneous document that was more poetic in nature could show similar doublings.
  - Arabic Al-Baqarah and Yusuf (consonantal): at or below chance. Arabic and Hebrew were including to look at languages that don't contain consonants and could show patterns that don't exist in long-form Latin languages because words could have multiple meanings.
  - Hebrew Psalms 113-150 / Hallel (liturgical Hebrew poetry): z = -2.87, p = 0.998- significantly SUPPRESSED below chance. The hypothesis here was that perhaps poetry in a language with a different structure could show similar doublings.
  - Hebrew Psalms full 150 chapters (19,662 words): z = -3.73, p = 1.000- even more strongly suppressed
  - Sanskrit Rigveda (complete, 135,279 words, IAST romanization): z = +3.04, p = 0.002- elevated above chance, though we flag this needs verification due to potential IAST sandhi encoding artifacts. Same hypothesis as liturgical Hebrew.

The Hebrew liturgical result is the one I find most interesting. Psalm 136 repeats the same phrase 26 times. Psalm 113 opens "הללו יה הללו". And yet the doubling rate is significantly BELOW chance — because real liturgical repetition always inserts at least one different word between repetitions of
the same phrase. "הללו יה הללו" is not "הללו הללו." That's also exactly what Carmina Burana does: "bibit hera, bibit herus-" same verb, different subject every time. Real repetition, whether poetic or liturgical, doesn't produce literal self-adjacency. Voynichese does, at ~2x chance, stationarily across the full manuscript.

The effect also survived a cross-section permutation test (p = 0.144 for the section gap — consistent with a single stationary process throughout), and a
local scramble test showing it depends on line co-membership rather than exact write-order.

2. Word-class ordering asymmetry — new to our analysis. Classifying words by their 2-character suffix (following Stolfi's grammar framework) and measuring directional ordering preferences between classes, both Currier languages show significantly asymmetric ordering: z = 4.72 for Language A, z = 4.77 for Language B, both p < 0.0001, robust to a frequency-matching control. Every other structural test returned null — this is the first positive,
grammar-suggestive result. I want to be careful: positional structure doesn't prove meaningful language. But it is something simple stateless generators don't produce by default, and the effect is equally strong in both Currier languages on the full validated corpus.

On the corpus: I used the complete FSG transliteration (Reeds 1994, from voynich.nu), validated word-for-word against the source file. In the process
we found that several Stars-section folios in our working corpus had been severely truncated (some folios had only 20-25% of their true content).
After rebuilding from the full authoritative file the corpus grew from ~14,000 to ~33,600 words (~88% of the manuscript). All findings were re-validated on the full corpus; the word-class asymmetry finding in particular changed meaningfully — what had looked like a gap between Language A and Language B disappeared entirely, with both now showing indistinguishable z-scores.

Happy to discuss methodology, findings, or criticisms. Hopefully it's clear what does and doesn't hold up (the Sanskrit result needs verification; an
earlier attempt to add corpus data produced fabricated text that was caught and removed before any analysis used it, and is documented in the repo). The GitHub repo has full documentation of every finding including the ones that failed or changed.

Thanks everyone, and looking forward to discussing.

.pdf   voynich_statistical_analysis.pdf (Size: 470.33 KB / Downloads: 16) ]]> <![CDATA[One funny word on f114r.30]]> https://www.voynich.ninja/thread-5867.html Thu, 02 Jul 2026 05:25:52 +0000 Jorge_Stolfi]]> https://www.voynich.ninja/thread-5867.html lkeede near the center of this image (f114r line 30):

.jpg   f114r.29-30.jpg (Size: 252.85 KB / Downloads: 182)
What do you see there? (No, this is not related to the BEEP theory.)

By my count, there are only three words that end with de in the whole Starred Parags section (SPS).  One is a qoteode, the fourth word on line 12 of f104r, a parag head line:

.jpg   f104r.11-13.jpg (Size: 306.98 KB / Downloads: 181)
Here the de apparently was the last thing that the Scribe wrote before pausing to recharge the quill.

The other de is on the first word of You are not allowed to view links. Register or Login to view. line 36, that happens to be a parag head too:

.jpg   f112v-36.jpg (Size: 292.19 KB / Downloads: 181)

The lkeede instance at the top of this post occurs near the end of a largish parag (f114r.28--31) There is no sign of quill recharging there, but the next three words aiin oeedaiin qoaiin seem to be a bit more crooked than usual.

There is no lkeedaiin or lkeedo in my transcription of the SPS, but there are two lkeedar, one lkeedal, one lkeedol, two lkeedain,one lkeed,  and a lot of lkeedy.

The digraph de occurs quite a few times in the middle of words, but mostly in groups dee or deee. In my alphabet/word model, those cases would be parsed as a {d} "element" followed by an {ee} or {eee} element, which is allowed.  Whereas de before any glyph other than e (11 occurrences in my SPS) would have to be parsed as a single element -- a d with an e modifier -- which is forbidden.

All the best, --stolfi]]> lkeede near the center of this image (f114r line 30):

.jpg   f114r.29-30.jpg (Size: 252.85 KB / Downloads: 182)
What do you see there? (No, this is not related to the BEEP theory.)

By my count, there are only three words that end with de in the whole Starred Parags section (SPS).  One is a qoteode, the fourth word on line 12 of f104r, a parag head line:

.jpg   f104r.11-13.jpg (Size: 306.98 KB / Downloads: 181)
Here the de apparently was the last thing that the Scribe wrote before pausing to recharge the quill.

The other de is on the first word of You are not allowed to view links. Register or Login to view. line 36, that happens to be a parag head too:

.jpg   f112v-36.jpg (Size: 292.19 KB / Downloads: 181)

The lkeede instance at the top of this post occurs near the end of a largish parag (f114r.28--31) There is no sign of quill recharging there, but the next three words aiin oeedaiin qoaiin seem to be a bit more crooked than usual.

There is no lkeedaiin or lkeedo in my transcription of the SPS, but there are two lkeedar, one lkeedal, one lkeedol, two lkeedain,one lkeed,  and a lot of lkeedy.

The digraph de occurs quite a few times in the middle of words, but mostly in groups dee or deee. In my alphabet/word model, those cases would be parsed as a {d} "element" followed by an {ee} or {eee} element, which is allowed.  Whereas de before any glyph other than e (11 occurrences in my SPS) would have to be parsed as a single element -- a d with an e modifier -- which is forbidden.

All the best, --stolfi]]> <![CDATA[Generators and Positional Tendencies]]> https://www.voynich.ninja/thread-5864.html Wed, 01 Jul 2026 17:59:12 +0000 JoeyB]]> https://www.voynich.ninja/thread-5864.html
Starting right at the front of the VMS, the two paragraphs of folio f2r:


.png   positional_map.png (Size: 244.48 KB / Downloads: 137)

What's been noticed by the pros here:

1. Tavie's vertical impact: a line's opener avoids the glyph directly above it, and it's directional (o→q is common, q→o isn't; o-o is avoided but s-s is fine). You can watch it in the opener column, they keep changing all the way down.

2. Anton's opener order (from "Regaining the lost order"): openers advance through a set sequence, k y d ch o q s sh. Anton reads You are not allowed to view links. Register or Login to view. as k-d-q in both paragraphs (which you can see), and says paras tend to open with a gallows. The first paragraph here it runs k→d→q→[c]→o→s→sh.

3. Top row: gallows p/f cluster on the first line of a paragraph (Tavie, Currier, Feaster). Here that's p in ypchol/ypchaiin (line 1) and f in fodan (line 8).

4. Feaster's drift: big picture within a line, sh→ch, qo→o, k→t shift rightward, and word makeup drifts as you go down the paragraph.

5. Line ends: the last word tends to differ from mid words (Currier).

So my question: how does a generator that writes left to right, one line at a time, produce these, especially the vertical one? What ties right drift to opener order to being allergic to verticality… maybe it’s separate habits, coincidence, astral phenomena…





// FWIW here is the rabbit hole of other threads I didn’t want to blow up with this question, and sources for positionality such I’ve been sorting through on this. This started with me just privately trying to aggregate rules and tendencies to harmonize over a nice pour of whiskey:

A. Other positional tendencies I left off the image:

benches (ch/sh) are rare as line-openers but spike later: You are not allowed to view links. Register or Login to view.

first words of lines run longer, second words shorter, Vogt 2012 (plain handwriting effect?): You are not allowed to view links. Register or Login to view.

long-range word correlations look like real topic structure, Montemurro & Zanette 2013: You are not allowed to view links. Register or Login to view.

q-start and m-end lines are anti-correlated (MarcoP and pfeaster in the vertical-impact thread below)

B. Sources for what's on the image:

vertical impact, Tavie: You are not allowed to view links. Register or Login to view.

opener order, Anton: You are not allowed to view links. Register or Login to view.

drift, Feaster 2022: You are not allowed to view links. Register or Login to view.

line as a functional unit, Currier 1976

C. Related threads: 

ch/sh interchangeability You are not allowed to view links. Register or Login to view.

copy/mutate ledger You are not allowed to view links. Register or Login to view.

hoax/generator debate You are not allowed to view links. Register or Login to view.

generators: Timm & Schinner 2019 (self-citation) You are not allowed to view links. Register or Login to view.

Cardan grille, Rugg 2004 You are not allowed to view links. Register or Login to view.

ReneZ 2021 write-up You are not allowed to view links. Register or Login to view.

Schinner 2007, lots here and the stochastic generator observation You are not allowed to view links. Register or Login to view.]]>
Starting right at the front of the VMS, the two paragraphs of folio f2r:


.png   positional_map.png (Size: 244.48 KB / Downloads: 137)

What's been noticed by the pros here:

1. Tavie's vertical impact: a line's opener avoids the glyph directly above it, and it's directional (o→q is common, q→o isn't; o-o is avoided but s-s is fine). You can watch it in the opener column, they keep changing all the way down.

2. Anton's opener order (from "Regaining the lost order"): openers advance through a set sequence, k y d ch o q s sh. Anton reads You are not allowed to view links. Register or Login to view. as k-d-q in both paragraphs (which you can see), and says paras tend to open with a gallows. The first paragraph here it runs k→d→q→[c]→o→s→sh.

3. Top row: gallows p/f cluster on the first line of a paragraph (Tavie, Currier, Feaster). Here that's p in ypchol/ypchaiin (line 1) and f in fodan (line 8).

4. Feaster's drift: big picture within a line, sh→ch, qo→o, k→t shift rightward, and word makeup drifts as you go down the paragraph.

5. Line ends: the last word tends to differ from mid words (Currier).

So my question: how does a generator that writes left to right, one line at a time, produce these, especially the vertical one? What ties right drift to opener order to being allergic to verticality… maybe it’s separate habits, coincidence, astral phenomena…





// FWIW here is the rabbit hole of other threads I didn’t want to blow up with this question, and sources for positionality such I’ve been sorting through on this. This started with me just privately trying to aggregate rules and tendencies to harmonize over a nice pour of whiskey:

A. Other positional tendencies I left off the image:

benches (ch/sh) are rare as line-openers but spike later: You are not allowed to view links. Register or Login to view.

first words of lines run longer, second words shorter, Vogt 2012 (plain handwriting effect?): You are not allowed to view links. Register or Login to view.

long-range word correlations look like real topic structure, Montemurro & Zanette 2013: You are not allowed to view links. Register or Login to view.

q-start and m-end lines are anti-correlated (MarcoP and pfeaster in the vertical-impact thread below)

B. Sources for what's on the image:

vertical impact, Tavie: You are not allowed to view links. Register or Login to view.

opener order, Anton: You are not allowed to view links. Register or Login to view.

drift, Feaster 2022: You are not allowed to view links. Register or Login to view.

line as a functional unit, Currier 1976

C. Related threads: 

ch/sh interchangeability You are not allowed to view links. Register or Login to view.

copy/mutate ledger You are not allowed to view links. Register or Login to view.

hoax/generator debate You are not allowed to view links. Register or Login to view.

generators: Timm & Schinner 2019 (self-citation) You are not allowed to view links. Register or Login to view.

Cardan grille, Rugg 2004 You are not allowed to view links. Register or Login to view.

ReneZ 2021 write-up You are not allowed to view links. Register or Login to view.

Schinner 2007, lots here and the stochastic generator observation You are not allowed to view links. Register or Login to view.]]> <![CDATA[Groups and weights of symbols]]> https://www.voynich.ninja/thread-5863.html Tue, 30 Jun 2026 20:23:38 +0000 ololololo]]> https://www.voynich.ninja/thread-5863.html I don't have any statistical research on hand, so I can't justify it with scripts or tables. I've determined all of this, let's say, "by eye".
Instead of using linguistic concepts such as "prefix," "root," "suffix," or models like "Crust-Mantle-Core," I decided to try to represent positional patterns in terms of character ratios (identifying which characters are "bigger" than others and assigning weights accordingly).
Here's what I have (sorry for the crooked drawing):

.png   groups.png (Size: 150.73 KB / Downloads: 38)
The symbols are grouped conditionally. The largest symbol here is o (ignoring q), and the smallest is y (it appears 23 times after n, which is smaller than even a conventional unit!)
The unit group is special in that it is difficult to define internal relationships. I can assume that r is greater than i based on 18 examples where r precedes i. d is less than e, because I believe that a = ei. But if we consider a as a separate symbol, it will be in the same group.
Some remarkable ratios:
1). ch's more than a dozens of them, but fewer gallows. Also, ch's more than o.
2). x is more than a dozens
3). In the category with the symbol q, I can enter c, which has the property of standing in front of gallows (I mean cases like ct, not cth)
4). It is difficult to determine the weight of s, but I assume that it is the largest in the dozens group, as well as more than ch.

I don't know what good it will do, but I did it Angel Well, at the very least, it perfectly demonstrates the peculiarity of the Voynichese letters. Since they can easily be assigned weights, maybe they're not letters at all, but something else... Maybe they're numbers... Maybe even You are not allowed to view links. Register or Login to view.?

What do you think?]]> I don't have any statistical research on hand, so I can't justify it with scripts or tables. I've determined all of this, let's say, "by eye".
Instead of using linguistic concepts such as "prefix," "root," "suffix," or models like "Crust-Mantle-Core," I decided to try to represent positional patterns in terms of character ratios (identifying which characters are "bigger" than others and assigning weights accordingly).
Here's what I have (sorry for the crooked drawing):

.png   groups.png (Size: 150.73 KB / Downloads: 38)
The symbols are grouped conditionally. The largest symbol here is o (ignoring q), and the smallest is y (it appears 23 times after n, which is smaller than even a conventional unit!)
The unit group is special in that it is difficult to define internal relationships. I can assume that r is greater than i based on 18 examples where r precedes i. d is less than e, because I believe that a = ei. But if we consider a as a separate symbol, it will be in the same group.
Some remarkable ratios:
1). ch's more than a dozens of them, but fewer gallows. Also, ch's more than o.
2). x is more than a dozens
3). In the category with the symbol q, I can enter c, which has the property of standing in front of gallows (I mean cases like ct, not cth)
4). It is difficult to determine the weight of s, but I assume that it is the largest in the dozens group, as well as more than ch.

I don't know what good it will do, but I did it Angel Well, at the very least, it perfectly demonstrates the peculiarity of the Voynichese letters. Since they can easily be assigned weights, maybe they're not letters at all, but something else... Maybe they're numbers... Maybe even You are not allowed to view links. Register or Login to view.?

What do you think?]]>