The Voynich Ninja

(24-09-2025, 02:31 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.that "rot" key was not followed by the Painter, and the letters look rather awkward.


So here is another theory: when the VMS Scribe was copying plant parts from another herbal, which happened to have been created in a German-speaking area, he saw the "rot" color key on the stem of that plant, and --- not speaking German -- thought that it was a Voynichese label that had to be copied too.  So he did, striving to interpret the German letters as Voynichese letters...

Color annotations were sometimes ignored by painters. Examples from the Vermont herbal are discussed You are not allowed to view links. Register or Login to view.. That same comment also shows that color annotations on You are not allowed to view links. Register or Login to view. were followed. The painters who applied colors were often (usually?) different from the scribes who wrote the annotations, so what the painters did or did not doesn’t tell us much about the scribes, in my opinion.

Considering how small they are, letters in the “rot” annotation seem to me rather ordinary. Maybe one could argue that the downward serif at the top left of ‘r’ is unusually long, e.g. compared with ‘r’ types 62/63 from Derolez’s book. Concluding from this possible tiny difference that the scribe couldn’t read German and that he was trying to hammer the Latin alphabet into Voynichese seem to me two cases of  “non sequitur”.

In my opinion, it is much more likely that the scribe who wrote the color annotation “rot” was a German speaker.

[attachment=11494]

(26-09-2025, 02:08 PM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.In my opinion, it is much more likely that the scribe who wrote the color annotation “rot” was a German speaker.

Well, I think that the original crib manuscript, from which the VMS Scribe was copying plant parts, had been written by a German speaker for his German scribe.  Like the examples that have been posted, with "rot" keys near some flowers that were, in fact, painted red.

But the stem of VMS You are not allowed to view links. Register or Login to view. was not painted red (or any other color), even though that page does have some (rusty) red (applied with a broad quill rather than a brush).   I think that is already evidence against those letters being a color key by the VMS Author to the VMS Scribe (or to some Painter to be engaged later).

Note that there are only two or three such "color keys" in the VMS.  If they were indeed color keys, We would expect a lot more -- like, almost one key on every page with flowers (the part whose color is most important).

While the letters in VMS You are not allowed to view links. Register or Login to view. do resemble "r" "o" "t" from contemporary manuscripts, they are rather awkward. 

Hence my belief that the VMS Scribe misunderstood the key that he saw on the crib herbal as a Voynichese label that he had to copy.  He probably turned the drawing sideways and mis-interpreted those letters as Voynichese glyphs: namely Ch o se

An then the Retracer further messed up the situation by not turning the page, and re-interpreting the rotated Voynichese glyphs as Latin letters.  Which he possiby did understand as the German word "rot".

(And yes, those letters definitely were retraced.  You can still see the original trace of the o, just inside the darker trace, on the NW side.)

All the best, --jorge

(26-09-2025, 05:26 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.the stem of VMS You are not allowed to view links. Register or Login to view. was not painted red (or any other color), even though that page does have some (rusty) red (applied with a broad quill rather than a brush).  I think that is already evidence against those letters being a color key by the VMS Author to the VMS Scribe (or to some Painter to be engaged later).

[attachment=11498]

So you believe that the “rossa” (red) annotations in You are not allowed to view links. Register or Login to view. image 22 are not color annotations, since those parts of the plant were painted green. Those parts are painted red in the related You are not allowed to view links. Register or Login to view..
Same for the unpainted parts of the drawings from LJS 419 discussed You are not allowed to view links. Register or Login to view..

What do you think the annotations in the Vermont and LJS medieval herbals are?

From the illustrated manuscripts I have seen, these annotations are not terribly frequent. I guess they mean that the scribe didn’t have the time to do the painting immediately after tracing the drawings, or that he planned to hire a different painter at some later time.

When there are no annotations, coloring was probably handled differently, either it was left to the free choice of colors by the painter, or it was agreed upon verbally with the scribe or some other person (e.g. author or patron) directly instructing the painter, or the same person who did the drawings took care of paint.

Sometimes, when the painting did happen, the annotations were not considered. Either they weren’t understood, or they were just ignored: things didn’t go according to the scribe’s intentions. These cases can possibly tell us something about the painters, their relationships with the scribes, maybe the time of painting. But what the painters did tells us nothing about the scribes who wrote the annotations.

(26-09-2025, 05:26 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Note that there are only two or three such "color keys" in the VMS. 

You are not allowed to view links. Register or Login to view. and at least some of the isolated letters (LET in the table) could well be color annotations.
All the COL annotations appear to be on Scribe 1 pages.

(26-09-2025, 05:26 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.If they were indeed color keys, We would expect a lot more -- like, almost one key on every page with flowers (the part whose color is most important).

Why do you expect almost one on every page? Are your expectations based on color annotations from actual medieval herbals? The Vermont herbal doesn’t have color annotations on most pages: is this what you expect? How many color annotations do you expect for the whole Vermont herbal?

In the whole Historia Plantarum, Ms 459 of the Casanatense Library, I am only aware of the color annotations on You are not allowed to view links. Register or Login to view. (G for “giallo”). Most pages have no visible color annotations. Is this what you expect?

Thread split.

For anyone interested in the paleographic aspect of the color annotations, I wrote about this two years ago: You are not allowed to view links. Register or Login to view.

My conclusion was that You are not allowed to view links. Register or Login to view. and the color annotations show differences (r, g) but also similarities (p). Probably a different hand from a similar context.

(27-09-2025, 09:57 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.So you believe that the “rossa” (red) annotations in You are not allowed to view links. Register or Login to view. image 22 are not color annotations, since those parts of the plant were painted green. Those parts are painted red in the related You are not allowed to view links. Register or Login to view..
Same for the unpainted parts of the drawings from LJS 419 discussed You are not allowed to view links. Register or Login to view..
What do you think the annotations in the Vermont and LJS medieval herbals are?

Of course those examples are color keys.  Indications for future painters of how that part is to be painted.  And so was the "rot" on the stem of the plant in the herbal that the VMS Scribe was copying from.   

But the writing on the stem of VMS You are not allowed to view links. Register or Login to view. is not a color key.  It is the result of the VMS Scribe mis-interpreting a color key as a Voynichese label to be copied, and struggling to identify siddeways "r" and "t" as Vounichese glyphs.  And then the Retracer, much later, doing the opposite.  That is my claim.

(27-09-2025, 09:57 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.I guess they mean that the scribe didn’t have the time to do the painting immediately after tracing the drawings, or that he planned to hire a different painter at some later time.

Agreed.  The keys are needed only if the manuscript is left unpainted by the Scribe and the eventual Painter is not expected to have access to the colored original where the drawings were copied from.

I suppose that "herbal mills" like those that produced the Alchemical Herbals would sell unpainted manuscripts, leaving the painting to the buyers.  

I think that would make economic sense, since drafting the outlines is the part that required expertise, while the painting could be done by anyone, even a child.  Thus the Scribe should invest his time on drawing more copies, rather than on painting those he drew.

(27-09-2025, 09:57 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.Sometimes, when the painting did happen, the annotations were not considered.

Possible explanations for colors that disregard the keys are that the Painter did not understand the language, or that he did not understand the purpose of the keys, or that he did not consider it important to paint with the "correct" colors.

The Udine copy shows why the Vermont copy had a key: because those parts that look like leaves were in fact ... duhhh ... flowers? that should be red.   But maybe the Vermont Painter was not Italian and did not know what "rossa" meant.  ("Red" is "rot" in German, "rouge" in French, "vermeja" in Spanish, "červený" in Czech, ... In Portuguese  "roxa" means "purple" ...)  Or, most probably, he did not care.

(27-09-2025, 09:57 AM)MarcoP 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. and at least some of the isolated letters (LET in the table) could well be color annotations.
All the COL annotations appear to be on Scribe 1 pages.

And all those occurences are all in pages You are not allowed to view links. Register or Login to view. to f32r, namely quires 1-4.  That is when the the Scribe delivered those quires to the Author, and learned, red in face, what "color keys" were and that he was not supposed to copy them...  Or when the Author fired the Scribe and recruited another one...

But let's look at those 13 "color keys" anyway:

• f1v "A single 'g' under the paint of the second leftmost green leaf. The suggestion is that this indicates that it should be painted green."
  
• f1v "A less clear single letter, possibly a 'J', inside the lowest yellow leaf right of the stem. It is not clear that this is ink under the paint. It is hard to identify this as a colour indication."
    I agree that the explanation for the existence of the first glyph in that spot must involve a color key.  But there are several problems with "just a color key".  The letter "g" would make sense only if the Painter spoke English or German; but in German "g" could also mean "yellow" ("gelb"). Moreover that glyph is not really a "g" in any common handwriting; "g" is merely the Latin letter that is the least dissimilar from it.  One could say as well that it is a Voynichese picnic table x with a curly r-plume upside-down...  So I would guess that, like the Ch o se of f4r, that glyph is the result of the Scribe trying to read a color key on the crib herbal (that may have been written sideways) as a Voynichese label.  As for the "J", that is a false reading.  That trace is only part of the faint outline of a leaf shaped like the "spade" card suit. It could be accidental offsetting from some other drawing elsewhere, like the plant in the lower right corner of Pharma f99r, the second plant in the top row of f100r, etc.  Coincidentally, the leaves and root You are not allowed to view links. Register or Login to view. are cut-and-pasted from the bottom right plant of Pharma You are not allowed to view links. Register or Login to view. (or vice-versa?).  Or it could be a first sketch of the plant, that was later abandoned in favor of the current one.
  
• f2r "Voynich text under green paint of bottom right leaf (middle petal) The suggestion is that this could also be a painting instruction"
    Apart from being too long for a color key, why should this key be in Voynichese, while the others are in Latin script?  Rather, this could be a label tied to the text, or some faint strokes (possibly offsetted from another page) that were "restored" by the Retracer as a Voynichese label.
  
• f4r "A capital F in the rightmost flower" 
    Why should this color key be a block capital, when other keys are in cursive? And what would this "F" color be?  Seems more likely that, in the crib herbal this "F" was not a color key but a letter label of the part, referring to a legend.  Which the VMS scribe duly copied without thnking.
  
• f7r "Characters probably forming the word "rot" (German for red) under the paint in the left half of the root. Could be an additional colour annotation similar to f4r"
    It is hard to figure out what the letters are under the paint.  The "o" actually looks like the "impossible" combination of an i left stroke with a reverse-e right stroke, namely the anti-a or updown-a.  The "t" may be more like a C. And the "r" seems to have a plume, more like an n or r...  So again I would say that this is a case of the VMS Scribe struggling to read a "rot" color key in the crib as a Voynichese label, and copying it as such.
  
• f9v "Several characters in the top left flower under the blue paint: "por" in the top petal, "p" in the lower left petal and "r" in the lower right petal. Readings are tentative. These may also be colour annotations."
  
• f9v "An unclear letter of scribble between the top two petals of the top left flower. Meaning not clear."
  
• f9v "A single "g" to the right of the top right flower Seems to be another colour annotation."
    I see a lot more writing in the petals of the flowers of this page. e. g. a Voynichese word beginning with ro on the lower left petal of the top flower.  But again they don't look like Latin letters. They can be more easily understood as the Retracer partly mangling some faint Voynichese words written on those petals for some reason.  Or even the Retracer mistaking veins on the petals for letters.
  
• f20r "A "p" (or possibly an "r") at the top of the root system Could be another colour annotation." 
    That "r" was retraced in darker ink. If it is not a wholesale creation by the Retracer, it can be explained again as the VMS Scribe copying an "r" color key as a Voynichese r under a stub of a stem, which was re-interpreted by the Retracer as an "r".
  
• f28v "Some apparent symbols in the middle of the flower If this is writing, the script has not yet been identfied."
    Those are just stamens and pistils, no?
  
• f29r "An "r" in the higher part of the root system. Could be another colour annotation."
    That root was majorly botched by the Retracer. Take for example the second chevron from the top, just above the chevron that has the "r", and check its left and right ends.  So Bog knows what that "r" was originally, or if it was there at all.
  
• f32r "A "p" and what looks like a "v" or an "r" in the bottom right flower. There could be another character after the "p". These could be more colour annotations."
    Again, those don't quite look like Latin letters.  But they do look like what the VMS scribe could have tried to make from Latin letters written sideway, namely "duh, maye an r and and an h".  And, again, further mangled by the Retracer...
  

There is one interesting observation:
Both those annotations and the marginalia show ambiguous tokens that could either be latin letters or Voynichese glyphs - or a mix of both.
The marginalia even show the full spectrum from unambiguous letters to ambiguous intermediates to unambiguous Voynichese glyphs. Why?

While I am far from convinced by the retracing hypothesis, I struggle with an alternative explanation. A clever obfuscation? Or incompetence?

The color annotations look much more refined than the marginalia, especially 116v. If we assume the artist copied color annotations from a source image without understanding them, his writing even got worse later.
However, can we rule out these annotations were added later, after completion of the text but before painting? From the imagery I would assume that marginalia are contemporary with the rest of the imagery on the respective bifolios. But that does not have to be the case for the color annotations. Either these were made at a different time period or by a different person as the marginalia.

(28-09-2025, 12:44 AM)Bernd Wrote: You are not allowed to view links. Register or Login to view.The color annotations look much more refined than the marginalia, especially 116v. If we assume the artist copied color annotations from a source image without understanding them, his writing even got worse later

My guess is that those "anomalous writings" resulted from two distinct processes.  Both assume that, over a couple of centuries, the text and figures had faded to such an extent that the owner engaged another experienced Scribe (the "Retracer") to restore the book, by carefully retracing all the faded parts.

Process [A] applies to the so-called "color keys" found on some plants in quires 1-4.  From the list above, I think that only a few entries are actually of this class:

• The "g" on You are not allowed to view links. Register or Login to view. (but not the "J").
  
• The 'rot" on the stem of f4r.
  
• The "rot" on the root of f7r.
  
• (Maybe) The "r" on the root of f20r.
  
• (Maybe) The "r" on the root of f29r.
  
• (Maybe) The "p" and "r" on the flower of f32r.
  

The other entries of the list above have other explanations.  Process [A] assumes that the Scribe who did those quires was copying parts of plants from other herbals, possibly "alchemical", provided by the Author.  One of those "cribs" was in German and had some color keys -- possibly written sideways..  Not knowing German, and never having seen color keys before, he misunderstood them for Voynichese labels to be copied, and struggled to interpret the Latin letters as Voynichese glyphs.  Then the Retracer later re-interpreted those misshapen glyphs as Latin letters, and "restored" them as such.

Note that, while the original Scribe was skilled in the handling of quill and vellum, he did not seem to have any prior experience with illustrated manuscripts.  Thus I think it is quite possible that he had never seen manuscripts with color keys before.

Process [B] applies to the long texts on You are not allowed to view links. Register or Login to view. and on the top margin of f17r.  Those texts were originally in Voynichese, probably by the Author.  A couple centuries later, they had faded so much that one could not even tell whether they were Latin or Voynichese.  Then someone tried to reconstruct those lines assuming that they were Latin letters.  This Retracer (who may or may not have been the same Retracer who restored most of the book) was a "German" (Swiss, perhaps?) and thought he recognized German words in the faded strokes.  

In particular, this "German" restorer assumed that You are not allowed to view links. Register or Login to view. was a religious charm with cross signs, like the ones in the herbal transcribed by Marco. And maybe it was, but in Voynichese...

By the way, I somehow got the impression that similar charms can sometimes be seen on the last page of books of the time, presumably as protection against the Spanish Inquisition That NOBODY Expects or whatever.  Am I hallucinating again?

Anyway, that is the current state of my delusions. All the best, --jorge

The issue I have with the retracing hypothesis is that, no matter how diligent, the retracer should have missed some strokes. We do not see this. Also when overwriting Voynichese with latin letters, we obviously should see something underneath in the multispectral images. Even if the original pigment completely faded, some other components of the ink or binding materials should leave a signal somewhere in the spectrum. Which proposed ink would utterly fade from UV to IR?

Second, a thoroughly overwriting leaving no trace must be equivalent to the source or verbose. It does not explain single letters or the short phrases. Again, there is nothing in the multispectral images suggesting there was ever more.

Third, why retrace the marginalia in such unreadable way if you're obviously not sure? That makes no sense unless the goal was to make the marginalia unreadable. We cannot rule that out but I doubt it.

There's something that has always bothered me:
Throughout the VM there is one recurring theme. Style-wise everything appears to be very closely related, but not quite. We see that in:
.)Imagery in Alpha vs. Beta plants and Nymph evolution
.)Imagery parallels in marginalia drawings (Nymphs, goat)
.)Voynichese in different scribal hands
.)Marginalia and color annotations have similar and dissimilar letters

How can this 'close but not quite' stylistic overlap be explained? I fail to believe this is a coincidence as it spans the entire manuscript. And the more people we add (artists, scribes, marginalia writers, month and quire number writers, painters, retracers), the more improbable it gets.

(28-09-2025, 01:10 PM)Bernd Wrote: You are not allowed to view links. Register or Login to view.The issue I have with the retracing hypothesis is that, no matter how diligent, the retracer should have missed some strokes.

On the contrary, we see plenty of examples.  For instance, on f2r:
[attachment=11499]

(A,F) Original traces that were not retraced. (B,D) Retracing changed the shape of the petal from blunt triangle to sharp lance. (C ) Original outline visible under retraced one. (E,G) Retracer added petals sticking out from side of original corolla, whose outline was (H).

(28-09-2025, 01:10 PM)Bernd Wrote: You are not allowed to view links. Register or Login to view.Also when overwriting Voynichese with latin letters, we obviously should see something underneath in the multispectral images.

  

Original traces were retraced when and because they were too faint.  The retracing (the first round, at least) apparently used ink with same formula and color as the original.  Thus the original traces should not be more visible in the multispectral images than they are in the Beinecke 2014 scans. 

Quote:Even if the original pigment completely faded, some other components of the ink or binding materials should leave a signal somewhere in the spectrum. Which proposed ink would utterly fade from UV to IR?

The ink was almost certainly not iron-gall ink. (Yes, I know about the McCrone report. Saving that for another post.) From its color and the way it faded, it was probably a solid pigment suspension ink; like china ink, but with some sienna/ocher pigment instead of lampbrack.  That is, basically watercolor paint.  

For writing or painting on paper, such ink would require little binder, because the pigment particles would lodge among the paper fibers and that would provide enough resistance against rubbing, for the expected life of a paper document or painting.  (Traditional China ink uses no binder, just lampblack.)

Such ink is not suitable for writing on parchment, since the pigment particles would just sit on the surface, and would be easily rubbed off.  As they obviously did.   Even if the VMS Scribe added some extra binder, it would probably have been gum arabic or some similar organic glue.  Obviously he did not add enough binder, or the binder softened by humidity and decayed or rubbed off anyway.  

Even if some of that hypothetical binder remained, those compounds have no spectral signature that is distinguishable from the parchment. 

Quote:Second, a thoroughly overwriting leaving no trace must be equivalent to the source or verbose. It does not explain single letters or the short phrases.

The main retracing round was careful to follow what remained of the original -- particularly so on the text, less so on the illustrations.  Thus its impact on the transcribed text, while not zero, was quite small.  It did not intentionally add any words or glyphs, but changed quite a few y into o or a, Sh into Ch,  Ch into ee (or vice-versa), r into s (or vice-versa), etc.. It also turned quite a few normal glyphs into of "weirdos", by mis-reading what was left of them. In my estimate, he made at least 2-3 such mistakes per page, on average; more or less, depending on how badly the page had faded.

Quote:Third, why retrace the marginalia in such unreadable way if you're obviously not sure?

Because the alternative was to leave them almost  invisible and watch them fade even more.

And he would not be the last Voynichologist who was quite sure of his guesses. 

Quote:There's something that has always bothered me: Throughout the VM there is one recurring theme. Style-wise everything appears to be very closely related, but not quite. How can this 'close but not quite' stylistic overlap be explained? And the more people we add (artists, scribes, marginalia writers, month and quire number writers, painters, retracers), the more improbable it gets

Indeed, the style of the whole manuscript is remarkably uniform.  The differences between alleged "hands" are much smaller than the similarities.  So much so that at least one handwriting expert stated that text and figures were all the product of a single hand.  He may have been wrong, but that shows how similar the "hands" are. 

Yet the Riemann Retracing Hypothesis (RH) actually strengthens that observation, since some of the style variations can be explained as the result of retracing.   For instance, on the You are not allowed to view links. Register or Login to view. pages several nymphs have a distinctive "showercap" diadem, which is not seen on f72r .  But those "showercaps" are all obviously later additions.  Take them away, and the nymph style on f72 becomes more uniform (or rather with a more gradual and natural evolution, as the Scribe perfected his nymph-drawing skills).

All the best, --jorge