The Voynich Ninja

(13-12-2025, 07:21 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.As a parallel, there are examples of similar effects You are not allowed to view links. Register or Login to view.

I saw that manuscript.

Of course you see that manuscript as proof that scribes would sometimes use bad ink that, by some magical physicochemical process, changed abruptly from light brown to darker brown in mid-stroke.  And would not care about that.  For hundreds of pages.

You surely will not see that manuscript as proof that perfect retracing (with a pen 50% wider than the original) is quite possible.  Or that a scribe (much more skilled than the VMS one) could do that for hundreds of pages.  Or that someone would want to do that.

You seem to have a very low notion of the skill and patience of Medieval scribes. Again, look at the artistic manuscripts, like the Book of Hours of the Duc de Strawberry, and estimate the accuracy with which the artist-scribes put down their paint and ink strokes. 

In order to write a manuscript with the VMS level of quality with 1.5 mm letters, an accuracy of 0.1 or 0.2 mm would be enough.  But higher accuracy would be needed when producing a higher quality manuscript, to obtain professional-looking letters of that size.   Or when hired to restore a manuscript that had terribly faded over the previous centuries.  Definitely there were scribes who could do that.

As for that Basel manuscript, my explanation for those ink density variations is that the scribe wrote an entire page (or chapter, or book) with a lighter ink and a sharp pen that left a thin trace everywhere.  Possibly an ink that would let errors be wiped off without trace.  Then, after checking for errors, the same scribe went back with a broader pen and darker ink, and retraced all the text.  But only the broad strokes of each letter.  Except for a few cases, he left the parts that were supposed to be thin -- like plumes and tails -- in the original ink.  And maybe did another pass later.

All the best, --stolfi

(14-12-2025, 12:21 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.As for that Basel manuscript, my explanation for those ink density variations is that the scribe wrote an entire page (or chapter, or book) with a lighter ink and a sharp pen that left a thin trace everywhere.  ... Then, after checking for errors, the same scribe went back with a broader pen and darker ink, and retraced all the text.

A variant of that explanation that may make even more sense:  the first pass, in light ink and fine pen, was written by a senior scribe who was fluent in Latin and knew all the scribal abbreviations and other tricks of the trade.  Because of his skill, his time was valuable, so he wrote as fast as he could while keeping letter size shape spacing etc good enough.  Then, presumably after some proofreading, a junior scribe was tasked with retracing the text with a broader pen and darker ink.  He did not have to know Latin or what the abbrevs meant, and could do it as slowly as he needed to.

And then I believe that another scribe went through his work, and fixed a few letters here and there, that the junior scribe had mis-traced.  This accounts for the scattered letters in even darker ink.

But I don't think that either scenario (two passes by the same scribe, or by a senior/junior team) applies to the VMS.  For one thing, the Rt0 (original) traces of the VMS are not just written with lighter ink, they are faded to a variable degree, all the way to invisible. Also, the VMS Rt1 (restoration) scribe did not retrace every broad stroke, as in the Basel example, but left many original strokes, glyphs, and words without retracing -- presumably because they were still legible at the time.

By the way, the Basel manuscript is in paper, not vellum.  That may make a difference: places that are traced twice with the same pen and ink -- like when traces cross, or are retraced within a short time -- tend to get darker on paper, because the ink penetrates the fibers.  Not so on vellum.

All the best, --stolfi

(14-12-2025, 12:21 AM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.You seem to have a very low notion of the skill and patience of Medieval scribes. Again, look at the artistic manuscripts, like the Book of Hours of the Duc de Strawberry, and estimate the accuracy with which the artist-scribes put down their paint and ink strokes. 

Can do. I have found different records of the dimensions for the Book of Hours in question, some catalogues list it quite larger than the others, the smallest I've seen is:

Book of Hours 23.8 x 17 cm
Voynich MS 23.5 cm × 16.2 cm

Which makes it just a tad taller than the Voynich MS, but I will just match the height of the sheets exactly. So, in the examples below the  crops from the Book of Hours maybe should even be a bit larger than they are. Note that we are comparing VMS to an exquisite commissioned artwork performed by renowned artists probably using the best available tools and materials. This truly should be state of the art and the highest quality the money could get you.

First thing I noticed is how bad the paint work looks when zoomed in to the same scale as we use when examining the Voynich MS. Not as bad as in the Voynich MS, but still quite below what I would expect. The lines in the Book of Hours are much thinner and probably made with a different kind of tool. However, the mismatches, gaps, unsteady curves are quite obvious at this scale and I don't think the artist(s) who made this Book of Hours would qualify for the retracing job in the Voynich MS.

What do you think?

[attachment=12966]

More hallucinations in Voynichcolor™® from page f17r, using the Lazarus 2016 multi-spectral images! Featuring another possible example of "imperfect retracing" (B).
[attachment=12980]
This clip spans the left end of lines 1-3 of page f17v. \as usual, the text shows clues of original traces (Rt0), the first genenral restoration pass (Rt1) and at least one subsequent retouching pass (Rt2).
   
    The image is a composite derived from three multi-spectral images taken by the Lazarus team in 2016, with wavelengths 365 (UV), 625 (red), and 870 (IR).  The processing included scaling the samples in each image from integer range {0..maxval} to the float range [0_1]}, where {maxval} is the sample value seen on the Spectralon sample included in the image. Then for each pixel, the 12 samples from the 12 narrow-band images obtained with both light sources (files "MB*" and seq numbers 001 to 012) were scaled so that their average value was 0.4. (This step largely eliminates variations from pixel to pixel due to vellum warping and fine-scale relief.) Then the images with the three selected wavelengths above were combined as the Y, U, and V channels of an image in YUV color space, which was then  converted to RGB.  At various steps the images were normalized so as to achieve maximum contrast in the area covered by writing anda bit of background around it.
   
    In this image, original traces are dark but fading blue, Rt1 traces are purple, and Rt2 traces are red with purple spots.  The colors are modified somewhat in areas (X) where the vellum color is affected by bleed-through of the green paint on the other side, f17v.  In those areas, all trace colors on this image are shifted towards red, so that Rt0 looks ike Rt1 elsewhere, Rt1 looks like Rt2, and Rt2 is bright red.
   
    (A) The loop of the f is mostly original (Rt0). (B) The plume is Rt0 except for a semicircular imperfect retracing by Rt1. (C,D,E,F,G,H,I,J) The body of glyph and the beginning of the tail are Rt1 or Rt2, but the distal half of the tail is still Rt0.  (K) The body of this glyph is Rt1 or Rt2, but the whole tail is still Rt0 and faint, so it could be transcribed as a. (L) Most of the plume is Rt2, but the tip is Rt0.
   
    (N) This glyph, originally o, was slightly deformed by Rt1 so that it could be confused with a y with faded tail.
   
    (P,Q) The C of the Sh is Rt2, while the h is Rt1. (R ) The C of Ch is Rt2, the h is Rt1. (S) The body is Rt2, and the plume was wholly retraced by Rt1 or Rt2. (T) The body and half of the plume of this r is Rt2, the rest of the plume is Rt1, traced the wrong way (clockwise). (U,V) Bit of Rt2 on Rt1 glyphs.

Here is the same area from the Beinecke Library 2014 scans:
[attachment=12981]
Frankly I don't see much gain in using the multi-spectral images here.  Al the details I can see in the pseudocolor image I can see also on the BL image.  The resolution is a bit better, and maybe it will be easier to automate the analysis with the Bayes method I described before.  We will see...

All the best, --stolfi

Thanks for them images and clips!

(15-12-2025, 09:18 AM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.The lines in the Book of Hours are much thinner and probably made with a different kind of tool.

I wonder what that could be.  

I suppose that in theory one can make a quill pen with arbitrarily fine trace.  It should be more fragile and should wear out faster, but if the payment is high enough...

Quote:Note that we are comparing VMS to an exquisite commissioned artwork performed by renowned artists probably using the best available tools and materials. This truly should be state of the art and the highest quality the money could get you.

Definitely.  

The original VMS Scribe was much below this caliber, and must not have had the same level of financial and professional motivation.  He did not even bother to keep the baselines straight. I would even guess that he had reasonable experience writing on paper, but not on vellum.  

Quote:First thing I noticed is how bad the paint work looks when zoomed in to the same scale as we use when examining the Voynich MS. Not as bad as in the Voynich MS, but still quite below what I would expect.

The human eye-brain uses only the lightness component of color to recognize shapes, including letter shapes and ornaments.  It records color proper with much lower spatial accuracy.  This fact has been exploited by some modern painters, by early color magazines and newspapers,  analog color TV, JPEG and MP4, ... If there are splashes of the right color at roughly the right place, the precise edges of the patch don't matter -- the eye will not see them, as long as there are strong light/dark edges (like an ink outline) with the correct shape.

Quote:However, the mismatches, gaps, unsteady curves are quite obvious at this scale and I don't think the artist(s) who made this Book of Hours would qualify for the retracing job in the Voynich MS.

Well, the smallest tendrils in the clips you posted are about as wide overall as a single stroke of the VMS clips.  To me, it looks like the Book of Hours artist had very good control of the pen, to less than 0.1 mm accuracy.  

Thus I think that he could definitely retrace the VMS original strokes so as to cover them completely, almost every time.

Note that, according to the MRT, the restoration Rt1 was commissioned by a later owner, who presumably had more resources than the original Author and thus could afford to hire a more skilled scribe (or scribal "factory").  Moreover, today the parts that were restored by Rt1 would be fainter than the parts that were not restored.  Thus,even if the Rt1 retraces were not quite perfect, and left a bit of Rt0 sticking out from the side, those defects would be very hard to spot today.

All the best, --stolfi

Quote:Meanwhile, here are some hallucinations of mine over f86v3

You may be interested in my thread about that page, Jorge:
You are not allowed to view links. Register or Login to view.

The "Earth" in the middle indeed looks unfinished.

Maybe the artist realised that he made a mistake and drawn Asia at the bottom and not the top and gave up with further drawing???

(15-12-2025, 07:52 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Well, the smallest tendrils in the clips you posted are about as wide overall as a single stroke of the VMS clips.  To me, it looks like the Book of Hours artist had very good control of the pen, to less than 0.1 mm accuracy.

The artist couldn't perfectly merge two lines together with a gap or step that would not be noticeable at this magnification. Even with strokes 3 times as wide an offset of 1/3 of the stroke width would be very easy to see. Also, the task of the original artist of the Book of Hours was much easier - when the artist had to merge two strokes, he could have merged at any convenient point along one axis (further or earlier down the line) so only one axis required high precision, while tracing a tight curve in the Voynich MS would require matching both dimensions at the same time. 

But even if the artist was able to achieve the same precision as in the Book of Hours, this is what the result would look like, probably.

[attachment=12995]

There should be many examples of similar or larger (since this is a harder task) misalignment of strokes on every page of the Voynich MS, if MRT was correct. And even that would require a true craftsman of high skill willing to dedicate months of work to the restoration of the Voynich MS. No restoration works are mentioned in the provenance of the MS. When would all of this retracement happen, between 1500 and 1600?

(15-12-2025, 09:07 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.You may be interested in my thread about that page, Jorge:
You are not allowed to view links. Register or Login to view.

Thanks!  My main interest in the post above was the retracing, and did not go into interpretation other than that comment.  Which was to reject the only thing we thought we knew about that page...

I really can't make much sense of f83v3.  The four "spouts" could represent the Four Winds, but their ends do not look much like winds.  Especially if my views about retracing are correct.  

The SE spout does not seem to be blowing or spraying at all. I believe that the sprinkles coming out of the SW one are a late addition by someone who interpreted the random scribble as the edge of a trumpet-like cone.   The NE spout seems to be pouring water rather than wind...

Quote:Maybe the artist realised that he made a mistake and drawn Asia at the bottom and not the top and gave up with further drawing???

Perhaps...  It may be relevant that this page is the last of the bifolio (the 2x3 foldout) when folded and stacked as it is now in the VMS.

But I believe that the "T" inside the circle is a later addition, because of the color of the ink; and that originally the circle was just empty.

The lines of the "T" are not as crooked as they would be if they were drawn entirely by hand (see for example f67r1), nor as straight as if they were drawn with a straightedge and the same ink attachment used in the compass (as the Scribe presumably did on f67r2).  

One should not use a quill with a straightedge, because the ink could easily be drawn under the latter.  (A common accident that I suffered a few times, with a steel pen.) 

So maybe that "T" was drawn with a normal quill, but with the fingers guided by a straightedge placed nearby but not touching the quill...

Also, whether the "T" is original or a late addition, whoever drew it may have been looking at that page upside-down, because of how the loose 2x3 "bifolio" would look like when unfolded and flipped over.

All the best, --stolfi

More retracing hallucinations over f86v3, aka "T-O Map [not!] and Garden of Eden  [?!]" or "The Four Winds [?]" (BL2014 image).
[attachment=12999]

   This clip spans the right (actually south) end of the paragraph of f86v3 that is adjacent to the E edge of the panel.  The clip above is rotated 90 degrees counterclockwise relative to the BL 2014 image.
   
    At least two and probably three levels of retracing are visible here: the original traces (Rt0), quite faint, surviving mostly as the tips of tails and plumes; the general restoration pass (Rt1) which here seems to be darker than the Rt1 of other pages, possibly because of imaging conditions or post-processing; and a scattered retouching (Rt2) in even darker ink.
   
    (A) The body and proximal half of the tail of this glyph are Rt1. The Rt0 trace is visible in the distal half of the tail and as a ghost parallel to the Rt1 trace. (B,C,D,E,F) The proximal half of the tail is Rt1, the distal half is Rt0. (G) The proximal half of each plume is Rt1, the distal half is Rt0. (H) This part of the p is Rt0, the rest is Rt1. (I) A bit of Rt0 visible below the bottom of the i. (J,K) Bits of Rt0 visible inside the upper loop of the glyph. (L) There is a bit of Rt0 trace at the top of the leg, above the start of the arm.  This f may have been a p, but the left loop was not restored except for bits near the top of the leg, and now has faded away completely.  Alternatively, the original was indeed a f but Rt1 drew the arm incorrectly as crossing the leg instead of starting at the top of it. (M) This ligature is probably Rt0.
   
    (N) This glyph, which is half Rt1 and half Rt2, is now ambiguous -- a or o.
   
    (R ) This glyph has the distal half of the plume in Rt1, the rest Rt2. (S) Isolated dark d assumed to be Rt2.


All the best, --stolfi

Repost of the annotated clip of the SE spout of f86v3, "The T-O Map".  This time with more annotations, including one case of imperfect retracing (F) and one good example of normal weight variation due to ink flow dynamics (S1-S8).
[attachment=13002]
This clip covers the distal half of the big "spout" rooted at the SE corner, including the bird nest and "stadium lights".  The clip is rotated 90 degrees clockwise relative to the normal reading orientation of the page.
   
    (A) Presumed original (Rt0) traces. (B) Presumed Rt0 traces.  Note that they cross the Rt1 outline of the "nest". (C,E) Presumed Rt0 outline of sprout. (D) Rt0 ribs of the sprout. (F) Rt0 trace of scallop, imperfectly retraced by Rt1 as (S7). (G) Bit of Rt0 (or maybe Rt1) not retraced by Rt2.
   
    (I) Edge of nest is Rt1, may not follow Rt0. (J) This part of bird may have been mangled by Rt1 and/or Rt2. Note the feathers of the wing crossing the outline of the nest, the backside of the bird in front of the wing, the doubly-traced hump, and the incongruous interaction between the upper filaments lines (A) and the edge of the nest,  See inset for possible original. (K) There is no Rt0 trace of these "stadium lamps"; they may have been added by Rt1.
   
    (Q) These circles are possibly Rt2.  ® Outline of the base of the spout by Rt2.  No Rt0 trace is visible; it may be an Rt2 addition.  The nebuly edge at NW, in particular, may have been created as the outline of the hatch lines above it.
   
    (S1-S8) Example of ink flow dynamics.  The Rt2 scribe started retracing these scallops from (S1) to (S2), tracing each scallop in the CCW sense.  He continued at (S3) and proceded up, row by row, right to left in each row.  At (S4) the pen began to run dry, but he continued in the next row from (S5) to (S6).  At that point he recharged the pen with ink and finished with scallops (S7) to (S8).


All the best, --stolfi