| Welcome, Guest |
You have to register before you can post on our site.
|
| Online Users |
There are currently 863 online users.
» 4 Member(s) | 855 Guest(s)
Applebot, Baidu, Bing, Google, Jimmy123, Wladimir D
|
|
|
| Objective statistics about VMS text |
|
Posted by: ThomasCoon - 17-12-2022, 01:16 AM - Forum: Analysis of the text
- Replies (10)
|
 |
Hello all, long time no see!
Please use this thread mainly to submit your objective (i.e. numbers-based) statistics about VMS glyphs, vords, and text.
If we gather all the facts together, we may be able to reverse-engineer the process that generated them. (Please feel free to speculate about that here too.)
Here are my statistics using Takahashi transcription:
- <l> appears 10,518 times in the VMS. In 54% of occurrences (5693 cases) it appears before <o>, and about 30% of times (3091 cases) before <a>.
So, 84% of the time that <l> appears in the VMS, it is either in an <al> or <ol> combination.
- <r> appears 7456 times. In 3244 cases (43.5% of the time), it is in an <ar> combination, and in 2820 cases (37.8%) it appears in <or>.
So, 81.3% of the time, <r> is either in an <ar> or <or> combination.
- <y> appears 17,655 times. In 14,515 cases, it is word-final.
So, 82.2% of the times that <y> appears in the VMS, it is word-final.
- There are 36,137 vords (tokens) in the Takahashi transcription. There are 14,515 vords with a final <y>.
Therefore 40% of vords in the VMS end in <y>.
- <d> appears 12,973 times. In 6834 cases (52.7% of the time), <d> is followed by <y>.
- When <dy> appear together (6834 times), it is word-final 88.8% of the time (6071 times)
- <k> appears 10,934 times, but in 6069 cases it is in the combination <ok>.
So <k> is preceded by <o> 55.5% of the time.
- <t> appears 6944 times in the VMS, but in 3856 cases it is in the combination <ot>.
So <t> is preceded by <o> 55.5% of the time.
- <q> appears 5423 times in the VMS, but in 5383 (99.3%) of cases it is word-initial.
- <q> appears 5423 times in the VMS, but in 5290 (97.5%) of cases it is in the combination <qo>.
- <qo> appears 5290 times in the entire VMS, but in 3116 (58.9%) times the next letter is <k>, and in 1130 cases (21%), the next letter is <t>.
So 80% of the time that <qo> appears, it is in a <qok> or <qot> combination.
- <ain>, <aiin> and <aiiin> appear word-initial over 500 times but are never once line-initial.
Your turn! You are not allowed to view links. Register or Login to view. is a text extractor.
|
|
|
| Quire and bifolio patterns for strangely-aligned text |
|
Posted by: pfeaster - 11-12-2022, 05:59 PM - Forum: Voynich Talk
- Replies (15)
|
 |
I was struck by Lisa Fagin Davis's remark during the recent Voynich Manuscript conference about the apparent organization of scribal production by bifolio. I knew that each bifolio tends to be associated with a particular scribe (or occasionally two), and that when a quire contains text by multiple scribes, their contributions tend to occupy different bifolios. In the past I think I've assumed this was a sign of bifolios being reshuffled at some point (since there's ample evidence of that happening in some cases): i.e., that Quire 4 contains a single Scribe 2 bifolio because that bifolio had originally been somewhere else. But the alternative idea that each bifolio was created separately, and that the results were subsequently assembled into quires, seems interesting, and I was curious to see if there were any other patterns (besides scribal attribution) that might point in this direction, or might shed further light on what was going on production-wise ("bifolio as a functional unit"?).
Here's one, perhaps: in both of the first two quires, strange alignments appear specifically on both recto pages of the outermost bifolio. In Quire 1, both You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view. have the final words of all their paragraphs aligned to (or near) the right margin: four cases on f1r, three on f8r. In Quire 2, You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view. each contain one paragraph starting with [p] that has a centered final line, along with one or two other paragraphs starting with [t] that don't. There are no other cases of similarly strange alignment to be found in either quire: just on the two recto pages of each outermost bifolio.
I think it's fair to hypothesize that pages with similarly strange alignment are somehow related to one another and that related pages are likely to have been written consecutively -- though of course neither point is guaranteed.
If so, that might suggest that:
1. The outermost bifolios of Quire 1 and Quire 2 were composed distinctly, in some sense, from the rest of their quires.
2. The recto pages of those two bifolios were also composed distinctly from the verso pages. (Note that the strange formatting of one recto page wouldn't have been visible as a model during the preparation of the other recto page.)
3. As bifolios were assembled into quires, somehow these two -- with similar characteristics -- both ended up on the outside of their quires, and right at the beginning of the quire sequence (or at least what's now the beginning).
If we follow this thread further into the manuscript, we continue to find other potentially interesting patterns in the locations of strangely-aligned text at the ends of paragraphs.
In Quires 3 and 4, each of the outer two or three bifolios contains one recto page with strange alignment, but alternating between the first and second recto page -- never both, and never the same one twice in a row. Then, further in towards the center of the quire, one additional bifolio in each of these quires instead displays strange alignment on one or both of its verso pages, rather than on a recto page.
Quire 3:
- [Alternating recto pages:]
- Outermost bifolio = You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. / f24r / You are not allowed to view links. Register or Login to view.
- Second bifolio in = f18r / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view.
- [Switch to verso:]
- Third bifolio in = You are not allowed to view links. Register or Login to view. / f19v / You are not allowed to view links. Register or Login to view. / f22v
- Innermost bifolio = You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. [no strange alignment]
Quire 4:
- [Alternating recto pages:]
- Outermost bifolio = f25r / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view.
- Second bifolio in = You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. / f31r / You are not allowed to view links. Register or Login to view.
- Third bifolio in = f27r / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view.
- [Switch to verso:]
- Innermost bifolio = You are not allowed to view links. Register or Login to view. / f28v / You are not allowed to view links. Register or Login to view. / You are not allowed to view links. Register or Login to view.
In Quire 5, the outermost two bifolios each have one page with strange alignment (f40v, f39r), this time with the "inner" case being on the second recto page and the "outer" case being on the second verso page.
Quire 6 is trickier to assess. In the outermost bifolio, f48r has a centered final line that seems to be avoiding an illustration, but the preceding line is also centered in a way that the illustration wouldn't have required. The second bifolio in has strange alignment on both You are not allowed to view links. Register or Login to view. and You are not allowed to view links. Register or Login to view. -- it's unusual for "mixing" recto and verso. The third bifolio in has strange alignment on f46v, of a peculiar kind (the seventh line of the paragraph is split into two halves with a gap between, with the second half establishing the left margin for the rest of the paragraph -- partly to avoid an illustration, but that wouldn't explain what's happening in the seventh line specifically). If these cases all "count" as unusual alignment, then this quire would follow a pattern of alternation vaguely like that of Quires 3 and 4.
There's no similarly strange alignment in Quire 7. Quire 8 is missing its three inner bifolios but has strange alignment on You are not allowed to view links. Register or Login to view. (a centered line at the end of one paragraph). That brings us to the end of the opening botanical section, which is as far as I'm going to go here.
Observations for Quires 1-8:
1. When a single bifolio contains two pages with strange alignment, they're always on opposite faces of the membrane (front/back), and they're usually on the same side (right or left), with the pairing f43r/f43v being the only exception to the latter rule. No bifolio has strange alignment on more than two of its pages, or on any two pages that would have been simultaneously visible during writing.
2. Within a quire, any verso pages with strange alignment usually appear further inward than any recto pages with strange alignment, with Quire 5 being the only exception (f39r is further inward than f40v).
3. Within a quire, two adjacent bifolios never contain a page with strange alignment in the same position (i.e., first recto, first verso, second recto, or second verso); and two pages with strange alignment never face one another.
4. Some of the bifolios with strange alignment are attributed to Scribe 1, and others to Scribe 2.
For the cases of "alternation" (e.g., the outer three bifolios of Quire 4), I wonder if there's any consistency as to whether the pages with strange alignment appear on the hair side or the flesh side of the vellum, and whether alternate bifolios might have been inverted to bring hair sides adjacent to hair sides and flesh sides adjacent to flesh sides. I understand it's a tough call to make as to which side is which, given how the vellum was prepared, but does anyone have any data about this?
Thanks for indulging this out-on-a-limb speculation -- I'm just trying to think of what other interesting consequences bifolio-level production might have had.
|
|
|
|
| Names of the divine beings of 82r |
|
Posted by: Ruby Novacna - 08-12-2022, 11:38 AM - Forum: Voynich Talk
- Replies (15)
|
 |
The image at the bottom of page 82r consists of three groups of nymphs representing characters from Greek mythology: two representations of You are not allowed to view links. Register or Login to view., each accompanied by a small nymph, and the group of three Charites or Graces.
The nymphs in the left and right groups have individual labels, the middle group of two nymphs has a single label.
Except the first label on the left «sororl» which could be Latin, the other four can be understood using the Greek dictionary.
The label for the first Aphrodite on the left, olko.ky EVA, would be Ανανη of ανανεω – come to the surface or of ανανεομαι – mount up – Aphrodite rising from the sea.
The label of middle Aphrodite, sokoly EVA, would be Ζωναια, epithet of Aphrodite, celestial goddess, member of ζωναῖοι, an order of divine beings managing the different zones of the Universe.
The labels of three Charites would be: - dolol – tola for θαλεια – rich, plentiful
- ol.aiin– awn for Αως – dawn, light of day,
- okeear–oneiar for ονειαρ – food, victuals.
The meaning of these labels corresponds well, in my opinion, to the meaning of the names of three Charites: θαλεια, Euphrosyne and Αγλαια.
The nymph Tola (Taula?), the oldest, would be θαλεια; the middle, youngest nymph, Aws (Awn) would be Aglaia- brightness, and the nymph on the right, Oniar, would be Euphrosyne- good food.
Could these labels help us to identify the language of the text, in your opinion?
|
|
|
|
| How much survives and how much is lost to history? |
|
Posted by: Mark Knowles - 07-12-2022, 06:18 PM - Forum: Voynich Talk
- Replies (20)
|
|
In my thorough research in ciphers from the time of the Voynich I have been very conscious of how much evidence has been lost to history. I have guessed that less than 1% of cipher keys and enciphered letters survive. This is on the basis of having a good idea of a sizeable portion of what survives; obviously I can't have seen everything and it is hard to quantify that of which I am unaware, however given that I have investigated the main European archives having such material I would be very surprised if I haven't seen at least 25% and at most 90%. I could again estimate the rough number of cipher keys/cipher ledgers and enciphered letters from around that time then in existence. These figures leads me to think that less than 1% of this original material survives to the present day.
I wonder how much of other kind of material survives. I would expect that a higher proportion of beautifully illustrated manuscripts survive. I wonder if enciphered letters were more likely to have been destroyed than other letters. So maybe enciphered letters are less likely than other documents to survive. Maybe cipher keys, proportionally, are more likely to survive than enciphered letters, however there also would have been many more enciphered letters than cipher keys.
I guess that crude estimates of the quantity of material surviving and the likelihood of a given document surviving can have some use in searching for documents be it in archives or online.
|
|
|
|
| From Voynich to the Beinecke |
|
Posted by: navalon - 30-11-2022, 03:39 AM - Forum: Provenance & history
- Replies (2)
|
|
At the upcoming Voynich Conference 2022, I will present a paper "From Voynich to the Beinecke," tracing the ownership of the VM which has not been precisely described previously. Dec. 1 at 9:30 US Eastern Time. I have now posted several relevant documents, including Voynich's will, a good copy of Ethel's will and documents from her probate file, some of which were not previously published. These are at You are not allowed to view links. Register or Login to view.
|
|
|
|
| Are the 1430s the most likely date range for the Zodiac drawings? |
|
Posted by: Mark Knowles - 24-11-2022, 05:15 PM - Forum: Voynich Talk
- Replies (110)
|
|
This was not a topic that I intended to look into at all a few weeks ago, but a claim that the Voynich pre-dates 1430 drew me in, so I have found myself looking at the Zodiac roundels and Koen Gheuens' excellent work on them. I tend to research only a few specific areas of the Voynich and this topic was not on my agenda. In fact it does feel like a bit of a distraction from my cryptographic researches. Nevertheless once I start something I tend to continue with it until when or if I feel I have reached an appropriate point to end. I am a little reluctant to raise this topic again as it got somewhat tense before, but it is an important topic and so I think should be raised given that I have given it some more thought.
I want to reiterate that Koen and others have clearly done a good job researching this topic, so that if I disagree with other people's conclusions it is not out of malice or lack of respect for their efforts, but purely determined by my thoughts on the evidence as presented.
As previously discussed and as Koen agrees there is no reason on the basis of the Zodiac clothing fashion to say that the Voynich manuscript could not date from the 1430s.
On his blog, Koen writes:
"Combining this information with the knowledge that the clothing of the VM Gemini, Virgo and archer were typical for the period 1400-1430"
Now if 1430 is not the end of the possible date range for the Voynich Illustrations then the logical implications of statements like this need to be re-examined.
This made me wonder whether the evidence in fact makes the Zodiac drawings more likely to date from the 1430s than other decades.
Having read Koen and Nick Pelling's posts on their blogs on the subject it got me thinking.
There were a few specific details that struck me.
1) The Zodiac Drawings appear to be exact copies
Koen points out how the Voynich Illustrator tries poorly to copy the shadings in a blue dress exactly indicating that the Illustrator was copying from drawings detail for detail.
Koen writes:
"What this suggests is that the VM painter was trying to faithfully follow an example, even though the required techniques surpassed his skill."
This would indicate that the Zodiac drawings were faithful and not altered copies of the source Illustrations.
On that basis it seems reasonable to think that the Voynich Illustrator would have copied the fashion exactly from the original manuscript and not updated it to fit contemporary fashion.
So why would someone alter the crossbowman's elbows? This certainly pushes the dating forward.
Of course if the central Zodiac drawings are exact copies from an earlier manuscript then those Voynich drawings could really date from any time after the inception of that fashion.
In fact given the central Zodiac drawings are copied whilst the other smaller Zodiac drawings around appear invented, it would seem that the central drawings reflect earlier fashion whilst the smaller ones may reflect contemporary fashion. This would imply that the smaller ones fashion are better suited to providing a contemporary dating.
However then one has to ask how much the original manuscript is most likely to pre-date the Voynich copies by 1 year, 5 years, 10 years, 20 years, 50 years. That is not an easy question to answer. Was this a culture of wanting to read the latest best selling manuscript or a culture where people happily pored over old manuscripts? Was someone more likely to be looking at a manuscript 1 year old or 10 years old? In theory one could calculate the estimated mean age of manuscripts consulted at that time. However in practice this would be very hard to do and not necessarily of much use.
2) Diebold Lauber drawings are the most like those of in the Voynich and Lauber's work dates from around 1427 onward.
From what I understand Diebold Lauber's manuscript of Buch der Natur(ca. 1440) is the closest known parallel to the Voynich Zodiac Illustrations; please correct me if I am wrong Koen. I have read that the workshop of Diebold Lauber produced manuscripts between 1427 and around 1467.
Now the obvious thing I wonder is if one of Diebold Lauber's works is where the Voynich Illustrator copied his drawings.
If the Diebold Lauber drawings are the most similar to the central Zodiac drawings then why not assume the date of the Voynich drawings is close to the date of the most similar Diebold Lauber drawings and why assume the original source is not a work of Diebold Lauber himself?
Lauber himself appears to have been known to copy older drawings in his manuscripts without making alterations for reasons of contemporary fashion. This allows plenty of scope for a later dating of the original Lauber manuscript from which the Voynich Zodiac drawings were copied.
Given that Diebold Lauber works date from 1427 to 1467 and given our carbon dating, does that not place the Voynich mostly likely in the 1430s?
It has been argued that Lauber and the Voynich author may have copied from a common source rather than the Voynich author direct from Lauber. However is that likely?
Are the similarities between Lauber and the Voynich too great to have been filtered through a previous author?
Why should an unknown earlier source be a preferable link to that of a known later producer with the greatest degree of similarity to the Voynich drawings?
One assumes the unnecessary complexity of this second source when the simpler option of the Voynich Illustrations being derived direct from Lauber seems the most obvious.
To quote Nick Pelling:
"the Voynich’s zodiac roundel drawings appear to me to have been copied (albeit fairly ineptly) en masse from a single (probably German) calendar of the late fourteenth or early fifteenth century. What’s different here is that it now seems quite likely to me that the particular calendar was from Diebold Lauber’s workshop."
After all this might suggest that the 1430s is the most likely range for the dating of the Zodiac Illustrations rather than being excluded, given the carbon dating pushes the date to before 1438.
Having looked more into this I wonder why it was ever thought the Voynich pre-dates 1430 based on the Zodiac fashion, that position seems to make little sense to me on closer inspection. I suppose it comes from the interpretation of quotes from a specialist or specialists as well as the mean date on the basis of Koen's collected of examples being 1415. It should be noted that specialists gave their estimate of the dates from which they believe the fashion originates not the date on which a manuscript containing such illustrations might have been copied.
As anyone who knows my research will know that for many years I have believed the author(s) spent a significant length of time in Basel at the Papal Council. This of course is not so far south of Haguenau, so a not unlikely place to find Diebold Lauber manuscripts. Note I made this association with Basel prior to any awareness of any kind of connection in the Voynich to North of the Alps, though I was of course aware of possible Northern Italian links. Obviously my narrative of a journey from North Italy, the Duchy of Milan, to Basel and backs accounts for both Italian and German influences on the Voynich.
So I naturally wonder if this theory is true then did the author see a copy of one of Lauber's manuscript whilst in Basel?
If one is claiming that the Voynich Zodiac drawings come from Lauber then the search for the earliest "surviving" Lauber manuscript with the most in common Illustrations with the Voynich seems a sensible one.
Has anyone made a list of Diebold Lauber's manuscripts that are known to survive and those that don't appear to have?
I saw a list that Nick Pelling mentions and I will probably discuss it with him, though I don't know how complete it is or what efforts have been made to systematically trace all of these manuscripts.
|
|
|
|
| Prague City University - The Voynich Manuscript - Research in Action |
|
Posted by: Scarecrow - 20-11-2022, 05:52 PM - Forum: Voynich Talk
- Replies (13)
|
|
Has this project been discussed here before? Dr. Stefano Cavagnetto's three year project on VMS sound curious..
"PCU research is centred on areas that help our communities. Sometimes this may involve finding practical solutions to real problems. And sometimes it may involve something that pushes the boundaries of our core disciplines and opens up insight into the mysteries of the world around us.
Uncovering a Medieval Manuscript
Dr Stefano Cavagnetto has uncovered just such a project as he is engaging on a three year journey in cooperation with the Italian Cultural Institute in Prague.
Centered around the mysteries of a medieval manuscript closely connected to Prague and Italy, he will lead a series of public lectures and seminars about the mysterious Voynich Manuscript."
You are not allowed to view links. Register or Login to view.
|
|
|
|
| Disambiguating Voynich Manuscript transliterations with word embeddings |
|
Posted by: Scarecrow - 18-11-2022, 07:04 PM - Forum: News
- No Replies
|
|
Jirka Lhotka, Francesco Salvi, Liudvikas Lazauskas
Department of Computer Science, EPFL, Switzerland
Abstract
Voynich manuscript is a 15th century document written in unknown language and script.
Consequently, reading and transliterating it to electronic form is equivocal and leads to ambiguities.
We present a way to disambiguate said uncertainties using word embeddings models, which achieves accuracies up to 86% on artificially corrupted texts of the same size.
You are not allowed to view links. Register or Login to view.
|
|
|
|
| Paula Findlens short article - Father Kircher’s Big Beautiful Books |
|
Posted by: Scarecrow - 18-11-2022, 06:58 PM - Forum: Provenance & history
- Replies (1)
|
|
"Publishing the German Jesuit polymath Athanasius Kircher (1602–80)
has always been a daunting enterprise. By the end of his life, Kircher pro-
duced a vast scholarly corpus. In 1634, he arrived at the Roman College from
war-torn German-speaking lands; by 1641, the Society of Jesus relieved him of
his professorial obligation to teach mathematics and Oriental languages to
focus on his publications. Almost fifty bulky tomes (mostly in Latin but occa-
sionally in German, Dutch, French, Italian, and English) bore his name. Most
were authored by Kircher; key disciples who assisted him and promulgated
his work (Gaspar Schott, Georg de Sepibus, Gioseffo Petrucci, and Johannes
Kestler) also published Kircherian books."
Full text: You are not allowed to view links. Register or Login to view.
|
|
|
|
|
