- Options
- Login
- Desktop Style
Options
[split] Jorge's Retracer Thread: darker ink, retracing of text and drawings
- [split] Jorge's Retracer Thread: darker ink, retracing of text and drawings
-
RE: [split] Retracer Thread: darker ink, retracing of text and drawings
Jorge_Stolfi > 05-05-2026, 08:54 AM
(05-05-2026, 04:15 AM)Charly Wrote: You are not allowed to view links. Register or Login to view.(what is the current consensus that the creator of the manuscript could have had connections with a trade route going through Constantinople?).
There is still no consensus about the origin and nature of the VMS. There are as many theories as there are researchers. Or even more than that.
Quote:I am attempting to interpret these images and would appreciate opinions on the following EVA defined glyphs, please (#5 is only present to show how faint parts of <qo> can become);
In general, my belief is that the Author wrote an original draft of the text on paper, with a rather cursive handwriting. The Scribe was taught the alphabet, but did not understand the language or the meaning of the text. He had some experience with a quill, and a passable handwriting, as if he had been an accountant or business secretary; but had no experience with drawing, writing on vellum, or writing books. And he was not motivated to produce a nice-looking result (as a professional scribe at a monastery would be). I think it is possible that he was doing the job as a favor for the Author.
So the Scribe often struggled to read the Author's draft, and that led him to substitute some glyphs by others that would look similar in sloppy handwriting. For instance, it seems that he would replace the ending in by r, and substitute d for k (but not t) or vice-versa.
Imagine someone trying to copy an text written with Latin letters but in cursive script and a thoroughly unfamiliar language. How often would he mistake an n for a u or a v, an y for ij, etc? On the VMS, these errors may have been as common as 1 in 20, or even more.
The Scribe would have also made many additional errors due to haste or fatigue. Like writing aiiin or ain instead of aiin, omitting an e, writing Ch as ee, etc. His glyphs were often deformed to the point that today we cannot be sure of what he meant to write. That is probably what created that stretched s on your image 6.
On some of those cases he would notice the mistake and go back, retracing the glyph so as to fix the mistake. That is part of my explanation for figure 4: he first drew the t with a very small right loop, then he went back and fixed it by drawing a bigger loop; but he had to re-draw the right leg closer to the left one.
Then, maybe a century of two later, there was that general restoration pass that retraced most of the text that had faded. Unlike the Scribe, neither the Restorer who did this job nor the Owner who commissioned it knew the alphabet, except from looking at the text (like we do now). So the Restorer often added more errors by "correcting" or "restoring" glyphs into the wrong ones, or even invalid glyphs.
This is my explanation for your figure 3: the Scribe originally drew the plume of the Sh out of place, almost on top of the second e, and then the Restorer "fixed" it by drawing a second plume in the right place. And also the second half of the explanation for figure 4: the Restorer tried to fix the three-legged t by filling the space between the two right legs,
That may also be the explanation for the qo with plume on figure 6. The two letters qo are usually a prefix, not an isolated word. I suspect that the original glyph was a Sh, which had faded do much that the Restorer mistook it for a qo -- leaving the original plume.
And then there must be errors made by the Author himself on his draft. In my origin theory, the Author had an imperfect knowledge of the language and was taking dictation from someone who may have been only partly literate himself. So he probably made many mistakes. (I lived for 13 years in the US, but I still cannot hear the difference between the vowels of "man" and "men"; to my ears they both sound the same, like the Italian or French "è". I don't think I would be able to distinguish the long and short vowels of German. Likewise there are many consonants of Arabic and other languages that I cannot distinguish. If I were to take dictation in those languages, I would get those letters wrong half the time...)
So I now believe that deciphering the VMS will be a painfully tedious and messy task because of all those errors. It will not be like cracking a cryptographic puzzle, where one just has to discover the algorithm, and then immediately one will get the whole plaintext without errors. It will be more like deciphering a lost language like Sumerian or Etruscan, where the speed of progress is measured in words per year...
All the best, --stolfi -
f116v: A hundred years of hallucination
Jorge_Stolfi > 7 hours ago
This post is a summary of the "Big Wipe" or "Big Spill" theory (BWT) for the text of You are not allowed to view links. Register or Login to view. (the "michiton" text). It claims that the original text was in the Voynichese script, but it was almost completely erased by a large water spill and then was incorrectly "reconstructed" as Latin-looking letters and vaguely Latin or Germanic-sounding words. Therefore, the text that we see today is not valid in any language and never had an intended meaning.
This theory would explain why all attempts to identify the language and infer the meaning in the last 100 years have completely failed.
In this post, @X denotes Voynichese text whose EVA encoding is X, while £X denotes glyphs on page You are not allowed to view links. Register or Login to view. that have been (or could be) read as X in the Latin alphabet.
All the best, --stolfi
You are not allowed to view links. Register or Login to view.
The "michton" text
Page You are not allowed to view links. Register or Login to view. of the VMS has only three small drawings in the NW corner (a nymph, a lamb, and an unidentified object) and four lines of text. At first sight, the latter seem to be an in an old handwritten form of the Latin script, except for two words of Voynichese script, in a mix of Latin and some Germanic dialect. The text is often called the "michiton" text, based on an early tentative reading of the first words of line 2.
The "michiton" text on f116v.
You are not allowed to view links. Register or Login to view.
On close inspection, however, one finds several glyphs that cannot be confidently identified as symbols form any known writing system. Those that do resemble Latin letters form mostly incomprehensible words.
These four lines are probably the part of the book's text that received the most attention over the last century, by far. However, all that effort has failed to produce a reliable reading of more that one or two words. Until recently, the purpose and meaning of the text remained unknown.
The words on lines 2 and 3 look vaguely like Latin or some Romance language. There is a cross or plus sign in most of the word spaces, and one word on line 3 can be read as "maria" with a cross on top. Because of these details, it is generally conjectured that those two lines are a Christian "charm", a magical protective formula (like the "Vade retro Satana" inscribed in some popular medals) that is to be recited while making the sign of the Cross at those marked points. On the other hand, the words of lines 1 and 3 look vaguely "Germanic", and many have concluded that they are parts of a recipe.
Interpretation
Recent observations have suggested a new explanation for the puzzling features of that. It turns out that this page had a singular history, quite different from that of other pages in the book.
Stains
In 2010 a set of images were taken from a few selected pages of the VMS, including f116v, using relatively narrow band illumination with various dominant wavelengths. Recently it was noted that, under ultraviolet light, one could clearly see the outline of a large water spill, covering a good part of the page.
The stain on page f116v. Ultraviolet reflected light image (MBMB365UV_007) from the ~2010 multi-spectral imaging session at Beinecke. (A) Edges of the big water stain. (B) Possible edge of chemically treated area.
You are not allowed to view links. Register or Login to view.
The extent of the spill is not clear in the upper part of the page, the area that holds the "michiton" text, but it quite possible that it spanned all the way from edge to edge.
Thus the stain on You are not allowed to view links. Register or Login to view. seems to be the key to the "michiton" mystery. What we see now is not the original contents of the page, but a largely unsuccessful attempt to reconstruct the same after it was mostly erased by the spill.
Inferred history
In more detail, this is the proposed sequence of events.
- Creation: The original Scribe (let's call him Rt0) wrote a three-line parag of Voynichese text at the top of this page (where we now see lines 2--4 of the text), leaving a blank margin 10-12 mm wide above it, typical of the other pages.
- Spill: Liquid water poured on the page, somewhere near the top, and flowed between folio You are not allowed to view links. Register or Login to view. and the back cover or some other surface, all the way to the bottom of the page.
When the spill happened, the page was somewhat slanted so that the water ran mostly towards the bottom of the page, without spreading out to the sides. While that page may have been facing up at the time, it seems more likely that it was face down, resting against some flat surface, such as the back cover of the book, a tabletop, a book rest, a cabinet wall, etc. Either way, the water was channeled by undulations of the vellum so that it split into two main branches, that joined agan near the bottom of the page.
The area covered by the spill is slightly darker than the surrounding vellum. This may indicate that the liquid was not water, possibly some beverage; or simply that the water picked up dust and pigment and spread it out move evenly over its extent. The boundary of the stain is smooth and is marked by a slightly darker line, suggesting that the water remained pooled over the lower part of the page long enough to begin to evaporate, forming a cracatristic "moraine" of dust and pigment.
- Wiping off: While the spill was still wet, someone wiped the liquid off from the top part of the page, above the line marked (B) on the image. To the person's dismay, this action wiped off most the original text too.
The two Voynichese words @aror @sheey at the start of line 4 presumably had been spared by the spill and survived intact. Of the rest, only a few scattered faint bits of ink remained, trapped in the tiny dimples of the vellum -- hardly distinguishable from random dirt spots and defects of the vellum. .
- Reparation: Someone -- not necessarily the same person who wiped off the stain -- then tried to repair the damage by retracing the lost text. However, this person (let's call him the "A Repairman", or "RpA") was apparently quite unfamiliar with the Voynichese script, and did not even realize that the lost text was in Voynichese. He therefore "reconstructed" most of the lost glyphs as Latin-looking letters, guessing many of them in an attempt to form words in Latin or some other language he knew. He apprently believed that the lost text was a prayer or charm and thus "reconstructed" some of the glyphs as the crosses that are often seen in those charms.
During this intervention, Repairman RpA also added a line of text along the N edge of the folio. There may have been a line of Voynichese text in that place before the spill incident, but there is no visible trace of it. It is possible that the Repairman got misled by stains formed from loose ink that collected in that area during the wipe-off. We note that the line lies above a sharp ridge of the vellum.
- Insect damage:At some point, probably after the Repairman RpA did his job, part of the page suffered significant """ + h.make_link(st, "insect-damage/clip.html", "insect damage", None, 0, 0) + """. The insects apparently entered through the holes now visible in the page. They may have been attracted by glue from the back cover that was dissolved by the water spill and coated onto the vellum. This damage had relatively little impact on the state of the page, thus we will not discuss it any further.
- Chemicals: There is some reason to suspect that, some time later, someone noticed faint remains of the original writing among those "reconstructed" glyphs. He would have applied chemicals to that area in an attempt to enhance those faint traces. (Maybe he was the same person who applied chemicals to f1r.) The evidence is that the vellum is visibly darker in that area, with a rather sharp boundary line (B).
If such tratment occurred, it must have assumed that the faint traces were iron-gall ink. This error probably resulted in those faint remaining traces being completely erased instead of enhanced. The treatment also seems to have caused the "reconstructed" writing of RpA to fade substantially.
Anyway, this hypothetical chemical treatment too would have had rela relatively little impact on the state of the page, thus we will not discuss it any further.
- Retracing: Then some person (RtB) -- maybe the same person who applied the chemicals, maybe not -- carefully re-retraced some of the glyphs that were produced by the Repairman RpA. most damaged. This "re-retracer" was was generally more careful and conservative than RpA; but he left some of RpA's traces as they were, or visibly deviated from them.
Indeed, it is quite likely that RtB was the same professional scribe Rt1 who did the global restoration of the whole VMS. But we will continue referring to him as RtB, just in case.
The "Latin" writing is not original
Even without considering the evidence of the spill, we can tell that the writing supposedly in Latin script was not original. Unlike the glyphs in the two Voynichese words, the letters of that text are extremely uneven and badly formed. See the detailed discussion below. In some cases ("m", "a", "s", "l", etc.), the same letter is written with two or more different shapes. Clearly the text was not written normally, but was traced slowly, letter by letter, with hesitation. As if tracing some now-invisible prior traces.
The original text was almost completely wiped out
It is well-established that the ink used in the original text and figure outlines (the "Rt0 traces") is not waterproof at all, and can be completely erased if it becomes wet. This property is proved by the "sauce" stain on page f103r, which erased the underlying ink so thoroughly that whoever tried to restore it could not read it any more. Thus, if the spilled water lingered over the writing for more than a few seconds, it must have softened the ink to the point that wiping the water off would have removed it almost completely too.
The Repairman had very little clues about the lost text
The situation facing the Repairman RpA of You are not allowed to view links. Register or Login to view. seems to have been quite different from that facing Rt1, the Scribe who did the general restoration of the whole manuscript, as detailed in other posts. On those other pages, the original writing had merely faded gradually over time, so that most of the traces were still at least barely visible. Thus Rt1 was able to recover practically all the text, missing or mis-reading only a few very faint strokes and skipping a few details (like plumes or tails) where the restoration would have resulted in ugly glyphs.
If the writing on You are not allowed to view links. Register or Login to view. had merely faded, like that on the other pages, the Repairman RpA should have restored most of the glyphs as Voynichese glyphs, even if malformed. The fact that he replaced most (but not all) of them by totally dissimilar Latin letters is evidence that the erasure due to the spill was much worse.
The Repairman had no notion of Voynichese
The scribe Rt1 who did the global restoration of the text on other pages had clearly had learned the general look of the Voynichese script and its alphabet. Thus, most of his re-traced glyphs were correctly formed -- although he probably was the source of most of the several hundred "weirdos".
In contrast, the Repairman RpA of You are not allowed to view links. Register or Login to view. was apparently quite unfamiliar with the Voynichese script, and thus he had no idea of what the lost glyphs could have looked like. Maybe he did not even realize that the original writing was not in Latin script.
Where a glyph had been completely erased, he presumably guessed a letter or two that would form vaguely plausible words in Latin or German with the other "reconstructed" ones.
The original text was not a charm
The crosses now seen on You are not allowed to view links. Register or Login to view. look very much like those that were drawn in or over the Chistian charms that are sometimes seen in other manuscripts, sometimes on the first or last pages. There, each cross instructed the reader to make the sign of the Cross when he reached that point in the recitation of the charm. The crosses on You are not allowed to view links. Register or Login to view. have lef many to the conclusion that those three lines were a charm too.
However, even if the original Rt0 text was in Latin letters, there is no reason to assume that it was a charm, or that those crosses were present in it. What probably happened is that RpA mistook the remains of some Voynichese glyphs, like @k, @t, or @q, as crosses; and from that he "deduced" that the text had to be a charm.
Indeed, details of three of the crosses -- (+3), (+5), and (+7) below -- suggest that those two were originally the heads of Voynichese @q glyphs. The Repairman RpA may have mis-identified the bits that remained of those two glyphs as crosses, inferred that the whole text was a charm, and inserted crosses in other places based only on this assumption.
There was a second retracer
There are a few bits of evidence that support the existence of the retracer RtB after the first clumsy repair job RpA. One is the apparent sharp distinction between two trace weights in the "Latin" writing. This variation does not seem to be due to pen re-charging.
Another evidence is two glyphs --- (+3), (+5), and (+7) below --- that show two traces, in those two distinct weights, that almost but not quite re-retracing (RtB) of the retracing (RpA).
The second retracer RtB also retraced the two Voynichese words on the third line. This retracing was careful and accurate.
Thus the goals and style of RtB are very similar to those of Rt1, the person who did the global restoration of most of the other pages. It is quite likely that they were the same person. In that case, it would follow that the "spill and wipe" accident and the clumsy recovery attempt RpA happened before the general Rt1 restoration pass.
A detailed look
In the rest of this page we look closely at the individual glyphs of the "michiton" text. At best,we can only guess at what the Repairman RpA thought he saw. Only rarely can be try to guess what was the original (Rt0) Voynichese glyph that he was trying to reconstruct.
Index to the glyphs from the "michiton" text.
You are not allowed to view links. Register or Login to view.
Collected glyphs from the "michiton" text.
You are not allowed to view links. Register or Login to view.
The Voynichese words
(V1..V9) These are the glyphs of the Voynichese words @aror @sheey. They may have been left untouched by RpA. The plume of the first @r (V2) and the tail of the @y (V8) are evidence for the second retracing RtB.
Recognizable "Latin" letters and symbols
(+1)..(+d) The crosses in lines 2 and 3. Cross (+1) is the first glyph on line 2, cross (+e) is the last glyph of line 3, and cross (+d) is above the word £maria, between the £A and the £r. All the other crosses are inserted in word or line breaks. Two word breaks on line 2 have no crosses.
Crosses (+2), (+5), and (+7) have a faint diagonal line connectind the two strokes across the NW quadrant, creating a shape like the digit 4. If they were "lazy pen" artifacts, due to RpA or RtB not lifing the quill off the vellum between the two strokes, they should close off either the SW or the NE quadrant instead. Thus they may be remains of the original Voynchese (Rt0) text. Namely, remains of originsl @q glyphs.
(a1)..(ac) These glyphs have been parsed generally as instances of £a. However, those with the distinctive "pointy hood", like (a2), (a3), (a5), (a6), and (a8) to (ac), seem to have been repaired first as £c, but then turned into £a by addition of the "hood". Only (a4) looks like it was repaired as £a right from the start.
Glyphs (a8) and (aa) are now rather faint. They may be RpA traces that were not retraced by RtB.
Glyphs (a1) and (n3) are here treated as two separate letter £a and £n, thus the first word of line 2 is implicitly read a £anchiton. Those two glyphs are often read as a single letter £m (M1). But then the first word would be £mchiton, not £michiton.
(o1)..(o9) These glyphs were generally read as £o, except (o2) on line 1, which has also been parsed as £u. However, the original Rt0 or RpA, now quite faded, was definitely a £o. The converson of that faint £o to £u was apparently the work of the second retracer RtB.
Glyph (o5) has been a longstanding puzzle because of the "bowtie". It may have been an @l originally.
(v1) This glyph was generally parsed as £v, and as such would be the only occurrence of that letter on ths page. It could be a malformed £r, although these are quite distinctive (see bwlow).
(t1)..(t6) These glyphs look like instances of £t in some handwriting of the period: namely, like a modern "c" with the top half less curved than the bottom half, protruding only slightly above a horizontal ligature dash. In this type of handwriting, that slight protrusion is often the only detail that distinguishes a £t from a £c.
(c1)..(c4) These glyphs have generally been interpreted as instances of £c in the assumed handwriting style. Their key features are a uniformly curved "c" that smoothly connects at the top with a ligature line strectching to the next letter.
They are very similar to the left half of Voynichese @ch, which probably was their original source. Occurrence (c2), in particular, is part of a £ce group that looks a lot like a @ch. Two occurrences, (c1) and (c4), are part of £ch groups; see (H1) and (H2) below. The remaning case (c3) is somewhat different and may be missing the ligature.
(h1,h2) These ghyphs are generally parsed as £h, because they both occur as clusters (H1,H2) which were assumed to be £ch. Their long curved descenders may have been originally the tails of @y glyphs. Their shapes are similar enough, although the second one has a "hook" at the end of the descender. Both seem to have been retraced by RtB except for the top half of the left vertical line, which is very faint and may be Rt0 material.
(e1)..(e8) These glyphs are gnerally parsed as £e. The key features that distinguish them from £t and £c are the absence of a ligature and an increased curfature at the top, that brings the top end towards the center of the glyph. Glyph (e7) even has a serif at the top that stressed that curvature. On the other hand, glyph (e3) has a "cedilla" -- a dot just below; it may be just an accident.
(d1)..(d5) These glyphs have been variously read as £s or £d, although they bear little resemblance to either. The reading £s is based on the "short" shape of that letter in some relevant handwriting styles when it occurs in word-final position. However, those "short s" letters are usually just as tall as an "o", whereas (d1)..(d5) are twice as tall. There are handwriting examples with "d" letters that look like (d2)..(d5), but this reading has generally been rejected because the presumed Latin words £multos and £portas would become £multod and £portad.
On the other hand, except for (d1), these glyphs all look pretty much like Voynichese @d.
(i1)..(ia) Most of these 10 glyphs have been traditionally parsed as instances of £i. Five of them -- (i2), (i6), (i7), (i9), and (1a) -- have the expected "dot". All have a mostly upright stem, except (i1) that leans to the right, and (i6) that leans strongly to the left. Glyph (i8) is rather faint and may be RpA trace; all the others look like they are RtB retraces.
The dots of (i6) and (i7) are stretched into diagonal line. Maybe they were @r glyphs in the original, and the current "dot" is the RpA intrepretation of what was left of the original plume. Those "dots" are also fainter than the respective bodies, suggestint that they are RpA while the bodies are RtB.
(r1)..(r8) All but one of these glyphs have been consistently read as £r, which in the assumed handwriting would be like a dotless "i" stroke with a poorly connected dot at its upper right. The exception is (r1), which is damaged and may have been meant to be an £n.
(m1)..(m5) These glyphs from lines 3 and 4 seem to have been intended as £m. There are two other possible instances of £m in line 2, shown as (M1) and (M2) below, but these are either ambiguous or have been generally parsed in other ways.
(n1)..(n6) These glyphs are generlly read as £n. Glyph (n3) has often been considered to one with (a1); see (M1) below. Another £n may be contained in the ambiguous glyph (M2) below. Glyph (n5) seems damaged and has a fuzzy short "reverse plume", whith three RtB dots, that merges with the "plume" of (c3).
(y1) This glyph would be naturally read as £y, but the conviction that the language was "Germanic" led to it being generally read as £n, as the ending "ren" seemed "Germanic". However, neither reading results in a valid word of any language.
(N1) This glyph occurs alone at the end of line 2, surrunded by crosses (+6) and (+7). It has been read as a large lowercase £n or as a uppercase £N. The latter would make sense as the placeholder for a person's name in a charm of curse formula directed to someone else; and that is presumably how RpA interpreted it.
(g1) This seems to be the only occurrence of the Latin letter "g" in the text. It is unusual because of the long "tail" in the descender (which apparently was retraced only in part by RtB) and a ligature line that cuts through the next £a (ac) and the next £s or £f (f4). Indeed the ligature, descender, and tail look like a single continuous "s"-shaped stroke, to which a "c"-like stroke was added to serve as the head.
(b1)..(b4) These glyphs have been generally parsed as £b. However, they are rather inconsistent: (b1) and (b3) have round tops, while (b2) and (b4) have sharp diagonal ones. Glyph (b3) is also open at the bottom. Like the £h glyphs (h1) and (h2), glyphs (b3) and (b4) have a faint section in the left vertical stroke, that may have been skipped by RtB. These glyphs may have been originally gallows.
(p1)..(p7) These glyphs were all definitely meant to be £p by RpA. The shape of the "head" is very charactristic and consistent. However, RpA may have traced the descenders of (p1) and (p5) only, and RtB re-retraced only half of the former. As a result, the descenders of (p2), (p3), (p6), and (p7) are now only faintly vsible in the UV image. They may even be original Rt0 traces, not even RpA. Those faint descendeers were missed by modern readers, so that all but (p1) and (p5) were generally read as £u or £v.
The only Voynchese glyph with a straight descender is @q; and the "heads" of (p1)..(p7) may be attempts to parse the heads of the common Voynichese @qo digraph.
(l1)..(l4) These glyphs are traditionally read as £l, but, like (b1)..(b5), their shapes are rather inconsistent.
(f1)..(f4) Glyphs (f2) and (f3) were always read as long-£s, and (f1) as an £f. Glyph (f4) was usually read as long-£s because then the words "£so £nim £gas £mich" could be read as "then take goat milk" in some unidentified "Germanic" dialect -- except that reading "£gas" as "goat" was quite strained, and "£mich" could be "milk" only if one assuemed the loss of an £l. Without that assumption, it is not possible to exclude the reading of (f4) as £f. The usual rules that determined where long-£s and short-£s should be used do not seem to hold on this page.
The original (Rt0) vestiges underlying (f2)..(f4), and possibly also (f1), apparently were just straight vertical lines, possible remnants of gallows and/or @q glyphs. In the case of (f2)..(f4), the horizontal stroke at the top which gave them the long-£s shape was provided by RtB; it is not possible to see whether RpA provided it first. On (f1), the whole visible glyph is RpA, possibly with some reinforcement by RtB at and below the cross-stroke. On (f4), RpA imperfectly retraced the vertical line, leaving bits of it still visible at the top and in the lower half.
(x1)..(x5) These five glyphs have been always read as £x. The first one is part of the word ending -£ox on line 1, while the other four occur on line 3 as part of the "charm" -- all four in the word ending -£ix.
The words "£pox £leber" have been conjectured to be "Germanic" terms meaning "buck's liver", and "£six" could be some "Germanic" dialectal form of "six". No plausible explanation has been given for the "£marix", "£morix", and "£vix".
Ambiguous, unrecognizable, or otherwise notable glyphs
(G1)..(G7) These glyphs have generally been read as compbinations of £t or £c with £o, £e,and other letters (see above). Gyph (G4) is usually considered unreadable. However, they are all sugestive of the Voynichese glyphs @ch and @che.
By the way, the ligatures in (G3), (G5), and (G6), and the left half of (G3) were not retraced by RtB. That ]of (G5), in particular, may be Rt0.
(M1) This gliph has been parsed into "Latin" letter in two or more ways. It was often read as an £m with an extra folded start, or as £mi, where the third "leg" would be the £i. This last reading was favored for a while since the £i seemed to have a dot. However, on the multispectral images that dot was revealed to be just dirt. Then a third reading, namely £an became more likely; see (a1) and (n3) above.
(M2) This string of minims could be parsed in many ways, including £mn, £mu, £nm, £um, £uni, etc. The parsing £mu was favored because then the word could be read as the Latin term "£multos" = "many". However, Repairman RpA probably intended the last letter to be £d. Thus the £mu reading is not much more likely than any other.
(C2,C2) These clusters have been discussed above as (a3)+(b2) and (a7)+(b3). They are reproduced here to note that they may well have been platform gallows in the original Voynichese text.
(H1,H2) The same observation applies to these clusters, discussed above as (c1)+(h1) and (c4)+(h2). They may have been originally gallows. However, the curved "tails" suggest Voynichese @y.
(D1) This isolated £o has always been a puzzle, especially because the row of dots. There are no visible Rt0 or RpA vestiges; the dots may have been created by RtB.
(Q1) This smudge at the right end of line 4, presumably RpA, looks too much like a question mark to ignore.
(S1)..(S4) These smudges may be vestiges of the Voynichese layer.
You are not allowed to view links. Register or Login to view. - Creation: The original Scribe (let's call him Rt0) wrote a three-line parag of Voynichese text at the top of this page (where we now see lines 2--4 of the text), leaving a blank margin 10-12 mm wide above it, typical of the other pages.
-
RE: [split] Jorge's Retracer Thread: darker ink, retracing of text and drawings
eggyk > 3 hours ago
(7 hours ago)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.(Q1) This smudge at the right end of line 4, presumably RpA, looks too much like a question mark to ignore.
I appreciate the write up. As for the supposed Q1 "?" symbol, I believe this is caused by the smudged/soaked stain from two EVA l symbols on 116r.
I overlayed the multiscpectral image of 116v over the normal scan in order to see much of the 116r text bleeding through. I then used that bleedthrough to align the 116r scan, and then imposed that over both.
-
RE: [split] Jorge's Retracer Thread: darker ink, retracing of text and drawings
Grove > 2 hours ago
One could almost imagine a bh instead of an sh at times. 8-) -
RE: [split] Jorge's Retracer Thread: darker ink, retracing of text and drawings
Jorge_Stolfi > 1 hour ago
(3 hours ago)eggyk Wrote: You are not allowed to view links. Register or Login to view.I appreciate the write up. As for the supposed Q1 "?" symbol, I believe this is caused by the smudged/soaked stain from two EVA l symbols on 116r.
Thanks! Yes, that would explain why the Repairman saw a question mark there.
But I think that he actually did draw the question mark: I see ink there, not just bleed-through from the two ls.
While vellum is translucent and lets one see a bit of what is written on the other side, unlike paper it is supposed to be impervious to water, and even slightly water-repellent. In the "ketchup" stain of f103r, it seems that the watery component of the sauce mostly stayed put, producing a stain with sharp edges; so that only the ink on the You are not allowed to view links. Register or Login to view. side was softened and erased. What got through to f103v, and then proceeded to You are not allowed to view links. Register or Login to view. and f104v, seems to have been the oily component of the sauce -- that did not affect the ink, and produced stains with fuzzy edges.
So the spilled water on You are not allowed to view links. Register or Login to view. did not get through the vellum and did not affect the ink on You are not allowed to view links. Register or Login to view. at all. Thus any ink on You are not allowed to view links. Register or Login to view. must have been placed on that page; not a grain could have come from f116r.
By the way, the spill obviously happened before insects created the holes along the top edge, because on You are not allowed to view links. Register or Login to view. there is no sign of water having come through them. As for the big hole on the vellum at the NW corner, the vellum is quite warped in that area, which must be why the water did not go through it either. And that must be also what spared the drawings in that corner, beyond the stitched gash.
All the best, --stolfi