In her recent (2020) Manuscript Studies paper “How Many Glyphs and How Many Scribes? Digital Paleography and the Voynich Manuscript“, Lisa Fagin Davis builds up to the conclusion (p.179) “The fact that all of these collaborative methods involve Scribe 2 may suggest that she or he was in charge of the project in one way or another.“
If this is the case, then I think the implicit question it suggests we ask is: what mistakes did Scribe 2 never make? That is, if Scribe 2 ‘knew what she or he was doing’ with Voynichese more than Scribes 1 and 3-5, we might sensibly expect Scribe 2 to make fewer scribal errors than the others. So, might we be able to use this prediction to tell good Voynichese (well, Currier B-ese, anyway) apart from miscopied Voynichese, hmmm?
The list of places where we can find Scribe 2 is as follows:
- All the Herbal B pages (apart from the f41-f48 bifolium, which was written by Scribe 5)
- The entire Q13 Balneo section
- One side of the nine rosette foldout (Scribe 4 wrote the other side)
- The first 12 lines of f115r (everything else in Q20 was written by Scribe 3)
One general problem with Voynichese is that – contrary to the wisdom of much of the Internet – it isn’t quite a game of two halves, i.e. a Currier A half and a Currier B half. Within those distinct variants, individual sections vary yet further: so, even though Q13 and Q20 are both ‘Currier B’, each one’s use of Currier B presents plenty of differences from the other. So if we are looking for differences, we have to be careful not to get caught up in the subtleties of how the (for want of a better word) style of Voynichese itself shifts between sections.
As a result, the two specific comparisons I think we should interested in here are the Herbal B pages (i.e. how does Scribe 5’s use of Voynichese differ from Scribe 2’s?) and the Voynichese on f115r (i.e. how does Scribe 3’s use of Voynichese differ from Scribe 3’s?). Let’s dive in and have a closer look…
Herbal B: Scribe 5 vs Scribe 2
The issue here is essentially comparing Scribe 5’s writing on f41 and f48 with Scribe 2’s writing on f26, f31, f33, f34, f39, f40, f43, f46, f50 and f55. Sadly, voynichese.com only offers a single filter of Currier A vs Currier B pages, which makes it not quite as useful as it might be (i.e. we’d like to do tests on [Herbal B + Scribe 2] vs [Herbal B + Scribe 5]). Maybe someone will add an LFD Scribe filter at some point in the future. 😉
But there is yet another dimension of difficulty to throw into the mix: transcription ambiguities. Because transcribers have quite a torrid time distinguishing characters (e.g. “a” vs “o” vs “y”, “cc” vs “ch”, “sh” vs “se”, and please don’t get me started on half-spaces vs spaces, *sigh*), we have to be careful we don’t mistake a transcriber’s whim for a scribal tell.
So the way I started was by grabbing the Takahashi transcriptions for f41r (Scribe 5) and f26r (Scribe 2), and comparing them really closely to high resolution images (on Jason Davies’ Voyage the Voynich website). My plan was to try to get a feeling for whether there was any visual evidence that indicated Scribe 2 was an author (i.e. who understood the internal construction of Voynichese) and the other just a dumb scribe (i.e. who was just copying what they saw).
However, I quickly found a fair few examples of what seemed (to my eyes) to be basic Voynichese scribal errors by Scribe 2.
- f26r line 2: second word looks like it should be “daiin”, but the first glyph is somewhat malformed
- f26r line 2: third word “adeeody” looks like a scribal slip for “odeeody”
- f26r line 2: free-standing word “lr” looks like a scribal slip for “ar”
- f26r line 3: word-terminal “-oy” looks like a scribal slip for “-dy” or possibly “-ey” (particularly in Currier B, though “qoy” is probably OK)
- f26r line 3: shapchedyfchy looks like a scribal slip for shopchedyfchy
- f26r line 3: penultimate word “saiin” looks like a scribal slip for “daiin”
By way of contrast, Scribe 5’s writing – though typically a little harder to transcribe – was generally quite clear, without any obvious scribal errors. So I would say that comparing these two pages (while only a relatively small sample) offers no obvious support to the notion that Scribe 2 might have had a more authorial understanding of Voynichese. On the contrary, it seems more likely to me from this that Scribe 2 was, well, just a scribe.
f115r: Scribe 3 vs Scribe 2
The first twelve lines of f115r (that Lisa Fagin Davies attributes to Scribe 2) present what look to me like yet more scribal errors by Scribe 2. For example:
- f115r line 1: “oechedy” (a hapax legomenon) looks like a scribal slip for “orchedy”
- f115r line 1: “oroiir” looks like a scribal slip for “oraiir”
- f115r line 3: the penultimate word “daar” (another hapax legomenon) looks like a scribal slip (possibly for “-dy ar-“)
- f115r line 3: the final word “oraro” looks like a scribal slip for “orary”.
- f115r line 5: the final word “ro” looks like a scribal slip for “ry”
- f115r line 12: the final word “choloro” should probably be “cholory”
(Incidentally, I should note that 12 out of all 13 instances of the word “ry” appear right at the end of a line [the other one appears right at the start of a line]. There is no shortage of patterns in Voynichese on all sorts of levels!)
Again, the remainder of f115r (attributed to Scribe 3) seems basically OK, so Scribe 2 again seems to be copying in the letters quite a lot worse than Scribe 3.
So… Scribe 2 was not the Voynich’s author, right?
From all this, it’s looking to me as though we can infer that Scribe 2 was not the author: or, more precisely, that Scribe 2’s errors seem consistent with the idea that Scribe 2 had no authorial level of understanding of the internal structure of Voynichese. Which, of course, would seem to be the opposite of what Lisa Fagin Davis’ paper suggested (if you read its conclusions in the strongest way possible).
However, I think this does imply something quite deep about the reliability of different sections of the Voynich, which is that some would seem to be less tainted by scribal errors than others. Though based on what is only a small sample, I suspect that Scribe 2 is a more unreliable Voynich scribe than both Scribe 3 and Scribe 5.
For a long time, I’ve been telling Voynich researchers that they should avoid treating the whole of the Voynichese corpus as if it were a single coherent text (because it isn’t): and that they should instead run their statistical analyses on individual sections, such as Q13 and Q20. However, because Scribe 2 wrote the entirety of Q13, I’m now revising that opinion: my particular concern is that Scribe 2’s copying errors (and I’ve only highlighted the errors I can see, there could easily be many others I can’t see) might well enough to disrupt any statistical studies.
Hence my recommendation going forward is that researchers should focus their decryption attempts on Q20, specifically excluding the top twelve lines of f115r (written by Scribe 2).
Why did Scribe 2 write the top part of f115r?
Might there have been a good reason why Scribe 2 wrote the top few lines of f115r?
Possibly. I’ve blogged a number of times about Q20 (which contains far too many bifolia to be a single quire), and how I think it may originally have been constructed as two separate gatherings Q20A and Q20B. The fancy gallows at the top of f105r looks a pretty good bet to have been the start of Q20A, and the current back page (f116v) similarly looks a good bet to have been the end of Q20B. I also wondered whether f104, f105, f107 and f108 may all have been cut from the same piece of vellum (an hypothesis which could at least be tested using DNA now).
(As an aside, I suspect that the seven dots on f105r imply that this marked the start of “Liber VII”. Just so you know.)
All of which would seem to point away from the (long-standing) suspicion that Q20 was written as a single monolithic slab, and instead towards the suggestion that Q20 / Q20A / Q20B might well have included separate sections. Might the first few lines of f115r have been written as the start of a section? Or might it even have been the end of a section, running on from a previous page (say, in Q8)? These are all no more than suggestions at this stage, make of them what you will. Possibly a nice risotto.
Q20 bifolio content notes
Finally: as Rene Zandbergen pointed out in 2016, the paragraph stars for the majority of f111r look to be fake. Yet the same seems true for the top half of f111v, as well as the bottom half of f108v, and the middle third of f115r; and there’s a paragraph star apparently missing from the middle of both f106v and f113r (though the latter of these two might possibly have just slid down the page). So we have to be extraordinarily careful when we try to draw inferences about the original section structure of Q20 based on the paragraph stars.
Here’s a brief summary for anyone trying to figure out the original nesting arrangement:
- f103: both recto & verso have no ‘x’ character (Tim Tattrie)
- – f104: recto has non-repeating star pattern (Elmar Vogt)
- – – f105: recto has fancy gallows at top, possibly start of Liber VII?
- – – – f106: verso has single paragraph star missing
- – – – – f107:
- – – – – – f108: recto has non-repeating star pattern (Elmar Vogt); verso has fake paragraph stars on bottom half
- – – – – – f111: recto & verso both have many fake paragraph stars (Rene Zandbergen)
- – – – – f112: both sides have a gap by the outside edge (possibly a copy of a stitched vellum tear, cf Curse 2006)
- – – – f113: recto has single paragraph star missing
- – – f114:
- – f115: recto has Scribe 2 writing at top (Lisa Fagin Davis) & fake paragraph stars in middle third
- f116: recto has no ‘x’ character, verso has michitonese + pen trial doodles
Very interesting post, Nick. Thank you.
The argument that the scribe who wrote the text didn’t know what she or he was writing is questionable in my eyes. Moreover there are 144 instances of “saiin“. This makes it unlikely that all 144 instances of “saiin“ are just scribal slips for “daiin”.
The first 18 lines on folio 115r are darker. On first view this gives the impression that this 18 lines must belong together.
What arguments did Davis present for her idea that only some of this 18 lines were written by Scribe 2? In her paper Davis only writes “the first twelve lines were written by Scribe 2” (Davis 2020, p. 177). There are no further details given.
Moreover this statement is contradicted by Lisa Davis own words. There is a youtube video from 2021 “The Voynich Manuscript with Lisa Fagin Davis” presented by the Beinecke Library at Yale. In this video at minute 68 Davis gives some details for Scribe 2 in Q20. The video shows an image of folio 115r. But instead of the first 12 lines all 18 darker lines are marked for scribe 2 (by using a red rectangle). Then Davis points to folio 115 verso instead of 115 recto and argues “if you go back and forth you can convince yourself that every line is written by a different scribe”. To make things even worse an image of folio 34v is used as illustration for scribe 2 on folio 115r.
Please see the video and decide yourself if you find Lisas hypothesis convincing.
1) I have not seen a confirmation of the 5 hands Lisa is postulating by a scholar I would take serious. Apart from the fact that variations in the script are easily explained by variations of a personal script in time, it simply does not make sense to me that there should be several scribes half of them don’t understand what they are writing, a single scribe and a bound up ledger of personal notes is a much easier and simpler explanation. Not to mention that there is a timeline in the illustrations, that somhow everyone ignores. The whole thing goes back to Currier, whose different statistis for Herbal A and B don’t mean different scribes and I suspect Currier was fully consious of that fact and has been misinterpreted
2) Dots are usually a signs for deletion of a string of letters, but in the case quoted here they seem to be simply ornamental
Torsten Timm: I wrote this post to try to test LFD’s conclusion, not my thoughts or your thoughts or even her (incomplete) presentation. (That’s for another day.)
While I agree that there are a fair few saiin instances, I was a bit suspicious about that specific one when I came to look at it. Perhaps I was overly suspicious there.
As to whether “the scribe who wrote the text didn’t know what she or he was writing” is questionable, I’d prefer to debate that in isolation from conclusions such as autocopying. That is, I think it’s a reasonable hypothesis to test against the evidence of what we see on the page.
If there are different scribes active here, and the writing we are looking at is cryptic, the history of people copying cryptic writing is that they often make mistakes (one or more per line of text is about normal).
So I personally think it is reasonable to use what we know about how Voynichese works to compare different error rates and/or different error styles between different scribes.
Helmut Winkler: if only appealing to simple-sounding solutions was enough to topple historical mysteries! We have overwhelming evidence of at least two hands (Currier’s Hand 1 and Hand 2), so anyone positing a completely solo effort has a palaeographic mountain to climb, in my opinion.
I have a number of reservations about LFD’s palaeographic reasoning (some of which I have raised with her, such as whether the EVA n shape can reliably be used as a palaeographic ‘tell’), but this post was about exploring the conclusion she drew about her Scribe 2, and viewing that same evidence from a very different viewpoint.
Quire 20 is a septenion.
The term isn’t usually in studies of western Christian manuscripts, because it is so rare there, but since the the definition is in the Lexicon of
Paul Hepworth and Karin Sheper, ‘Terminolology for the Conservation and Description of Islamic Manuscripts‘.
I suppose it’s less rare for them. Not surprising really. We’ve known for a century that the person who had the ms longest, and went to the greatest amount of trouble to discover more about it had been led to think the material gathered in the eastern side of the Mediterranean.
https://www.islamicmanuscriptconservation.org/terminology/alphabetical-index-en.html#jumptos
More on septenions.
A first quick hunt through Brit.Lib. and JSTOR turned up a manuscript which I think will cause pricked ears here.
“This manuscript consists of a single septenion. It has 22 lines to the page.
The text occupies 12.2 x6.4 cm.
fol. 5. Rynucius Poggio suo Oratori Eximio I felicitatem (in
red) Ille Rem optimam et sibi salutarem ….. (fol. I8). At in
uita nemini datur effugere fatum. (One line blank) FINIS.
fol. x8.’ Blank.
An unpublished letter of Rinucci da Castiglione to Poggio, with translations
of the Athenian decrees contained in the De Corona of Demosthenes. The
letter must have been written before I459, when Poggio died; probably before
I453, when he left Rome; and possibly much earlier still, as he was studying
Greek with Rinucci as early as 1425. See Voigt, Wiederbelebung des klass.
Alterthums, 1893, II, pp. 45, 84. The present copy might well have been
made about the middle of the century,
From: E.K. Rand, ‘A Harvard Manuscript of Ovid, Palladius and Tacitus’, The American Journal of Philology, Vol. 26, No. 3 (1905), pp. 291-329 (39 pages).
I’m sure one might find more.
Cheers
and finally on septenions.
More examples from Latins’ manuscripts listed in references for the entry ‘Septenion’ in French glossary at
Wiktionnaire: le dictionnaire libre
I take it that we accept the central bifolio has been removed. That’s how the quire is described in the Yale facsimile edition.
Peteb: No ideas on your so called conspiracy unless it’s a conspiracy consisting of you and ‘Good on you Gordon’ to downplay the part our new German immigrant scientists and engineers played in regenerating industrial growth after just having the scheisse beat out of them on their homefront. Also the pro nazi companies like your I. G. Faben combine proved to be the catalyst in creating post war prosperity under brands like Agfa, BASF and Bayer to name a few. No before we knock the krauts et al we gotta consider all the equally bad stunts we pulled on them like bombing population centres, condoning attrocoties undertaken by our victorious allies, or closer to home, interning thousands of harmless foreign and local born alliens in conditions equal to Changi under the nips. Fair go on old Jerry cobber.
I guess I should have spelled out the relevance of that manuscript.
The VMS has some quires that are quaternions – standard for Latins manuscripts, and for Greek manuscripts.
As you go down into the lower levels which, in the usual way would be the first in the stack, you find quinions – a little less usual in Latins’ manuscripts but not unusual in earlier Irish manuscripts.
Then you get to quire 20 – originally a septenion – which is very rare in Latins’ manuscripts.
Now, at Harvard, we have a group of texts, one of which is a septenion and this includes a letter to Poggio Bracciolini, usually credited with having invented the ‘humanist hand’.
It also includes 10 quinions – pretty standard for manuscripts from the eastern Mediterranean and these days usually associated with Arabic manuscripts though – as above – not unknown in earlier Irish manuscripts.
(I covered these distinctions for the Vms some months ago at Voynichrevisionist, referring readers to – among other sources, Adam Gacek, Arabic Manuscripts: A Vademecum for Readers (2009). As he says ( p.212) where he says, ‘the overwhelming majority of parchment quires in Arabic manuscripts consist of quinions (i.e. five bifolia, ten leaves)’.
So now the texts being copied for Poggio are ‘new’ sources for ancient Greek and Roman writings, but it looks as if when such works were recovered from the eastern side of the Mediterranean (or, say, more generally within areas where Arabic was a standard language), the copyists might also copy the format rather than break an original quinion, or even a septenion, into separate quires.
The Harvard ms (which by the way is digitised) doesn’t appear to have been made, or put together, in a hurry. But if, instead of these ‘rediscovered’ materials being brought into Italy, they had had to be copied in whatever situation they’d been found then you might find just what we seem to have in the VMS – a couple of scribes who knew what they were about, but one(?) ring-in, less able, who was there just to hurry things along and who then might well make mistakes.
I’m perfectly sure the VMS’ didn’t have one ‘author’ and that the authors of the original material were long gone when the Vms was made.
So if that parallel in quires’ structure allows us to posit a similar context in general terms (copying older works in the new ‘humanist’ style for someone interested in Gk as well as Latin sources) and that the curious use of quinions and septenions in both is due to the customs in place when the exemplars were made, then it balances pretty nicely and might lend support to some of Nick’s proposals.
And certain other and older assertions about the VMS.
Interesting read, Nick.
I do not understand this obsession with the structure of the quires, since one of the few things we know for certain about this manuscript is that the current structure of the quires is not the original structure of the quires and of the manuscript itself.
Those who made a septenion of quire 20 are the same people who wrote the western numbers on each folio, because, I’d like to point it out, without those late numbers nobody would suspect that was a septenion. That’s it.
Hi, Nick,
These are important observations, but you’re making a very different point than the one I was arguing. All of the scribes are guilty of making “mistakes” – that is, using Voynichese in unexpected ways that look like errors. Scribe 2 does indeed write many of these oddities. That observation doesn’t have any bearing on whether Scribe 2, or any of them, might have been in charge of the project in some way. I’m certainly open to other interpretations as to why Scribe 2 appears as a collaborator in several places, but the Scribe’s relative skill, or lack of, doesn’t argue against my original observation.
Lisa Fagin Davis: it is your conclusion (drawn from your palaeographic analysis) that Scribe 2 may have been a lead collaborator, but it is my conclusion (drawn from the specific types of scribal error that Scribe 2 seems to make) that Scribe 2 did not have (what one might call) a “creator’s knowledge” of how Voynichese works, or else he/she would surely not have made such basic mistakes.
I would agree that they are different points drawn from fundamentally different kinds of historical evidence, sure: but comparing and contrasting heterogenous sources is how good History works, in my experience. To what degree the two are commensurate or incommensurate depends very much on a whole load of secondary scenario theorising, far more than can be shoehorned into such a small margin as this.
Stefano Guidoni: the reason I included all the details about Q20 at the end of the post was because Lisa Fagin Davis’ observation about the Scribe 2 section at the top of f115r was new to me, and I didn’t have a single post listing (many of) the things a researcher would need to know in order to try to reconstruct Q20’s original quiration / gathering arrangement (e.g. Q20A + Q20B).
Um, um is this saying the book was written in a small job shop
by Greek scholars and scribes that fled Constantinople ?
Nick: I do not doubt that it is not an easy task to copy a text someone can’t read. I have doubt that it is possible to detect copying errors in a text someone can’t read. This is especially true if the writing looks cryptic. You argue that from your knowledge about Voynichese you can detect and correct scribal mistakes. But we didn’t have any evidence that text was indeed copied and since you didn’t give any criteria for your “corrections” your choices only depend on your ideas about Voynichese. Without any criteria given there is no way of validating your choices.
For instance you argue that “adeeody” must be a scribal slip for “odeeody”. But also a word “odeeody” would be unique. To you also “lr” looks like a scribal error for “ar”. But beside “lr” on folio f26r eleven more instances of “lr” exists. To you word-terminal “-oy” looks like a scribal slip for “-dy” or “-ey”. But words ending in “-oy” are not uncommon for Currier A. There are for instance 13 instances of the word “choy” and 9 instances of “qoy”. Since Currier A words can also occur in Currier B it is no surprise to find words ending like”-oy” also on a Currier B folio. To you “shapchedyfchy” looks like a scribal slip for “shopchedyfchy”. But a word “shopchedyfchy” would also be unique. To you even the common word “saiin” looks like a scribal slip.
How can you be sure that the corrections you suggest must be correct?
Torsten Timm: the post flagged that -oy words are uncommon in Currier B, so you already know the answer there. Similarly, you also know that every instance but one of “lr” appears on the end of a line, where “am” is extremely common. You also know that “ap” is rare compared to “op” (op occurs nearly 100x more often), so it’s not hard to work out why I suspect the former is likely a scribal slip.
You can see all this stuff too, probably better than me, truth be told. It’s not exactly a secret.
Not saying this is relevant. Not a theory. Just an historical-cultural note.
It is usually presumed, when speaking of Latin Europe, that a scribe will copy direct from an exemplar, but it was very common that a group of students-as-scribes would sit and take dictation. We have records that this was also the case in medieval Constantinople – and what’s more, it was done out of doors.
In copying the holy texts, in order that the possibility of error should be reduced to the absolute minimum, I am reliably informed that the custom in Jerusalem was to prefer scribes to be able to write, but not to read. The dictation would be read one letter at a time, and at the end of each verse (or perhaps each line?) the number of letters would be counted, and again at the end of each chapter. This prevented a scribe from anticipating the words and from duplications.
I’m not producing a theory about the VMS, but it might be worth considering along with the usual idea of directly copying from an exemplar.
PS – re your Rosicrucian posts which allow no comment – what sort of cipher did Elias Ashmole use? Did he invent his own?
Nick: You implicitly argue that Voynichese must be structural similar. But isn’t it likely that someone copying again and again words sharing structural similarities would also notice this similarities? With other words, isn’t it more likely that a scribe would correct “saiin” into “daiin” after he has copied numerous instances of “daiin”, “dain”, and “daiin” before?
Anyway, the idea that rare elements can be dismissed as mistakes is a very common one. Lisa Fagin Davis even argues that “qo” should be read as a single glyph since “q” is almost always followed by “o”. However in the context of Voynichese there is always a word or an element that is used more rarely. Therefore for Voynichese the argument that something is used rarely is not enough to dismiss it as an error.
For instance beside “daiin” (863 times) also “dain” (211) exists …
beside “saiin” (144) there is “sain” (68) …
beside “lor” (43) there is “lr” (12) …
beside “lr” (12) there is “ls” (10) …
beside “qody” (17) exists “qoy” (9) …
beside “qoy” (9) exists “oy” (6) …
beside “shok” (5) there is “chop” (2) …
beside “chop” (2) there is “chap” (1) etc.
“… the resulting network is connecting 6796 out of 8026 words (=84.67 %). … The longest path within this network has a length of 21 steps, substantiating its surprisingly high connectivity. …” (Timm & Schinner 2020, p. 4).
D.(iane?):
Re: Ashmole’s cipher, http://emlo-portal.bodleian.ox.ac.uk/collections/?catalogue=elias-ashmole says, “The catalogue comprises 764 letters… Many are written in Ashmole’s own cipher, but these were deciphered by C. H. Josten and published in plain English in C. H. Josten, ed., Elias Ashmole: His Autobiographical and Historical Notes, his Correspondence, and Other Contemporary Sources Relating to his Life and Work, 5 vols (Oxford: OUP, 1967).”
I don’t have a subscription to view the content, but from https://www.oxfordscholarlyeditions.com/view/10.1093/actrade/9780199670253.book.1/actrade-9780199670253-div1-5 Josten appears to discuss the cipher in the introduction to Vol 1.
A PhD thesis at https://scholarship.rice.edu/bitstream/handle/1911/87752/BROCHSTEIN-DOCUMENT-2015.pdf?sequence=1&isAllowed=y says, “Ashmole’s cipher is ‘identical with’ the one described in 1602 by Willis (d.1625) in the Art of Stenographie.158” The footnote references Josten.
P.S. Glad to see you blogging again. Global pandemics suck. Stay safe down under.
Just jumping in to respond to Torsten: my argument for [qo] being considered one glyph instead of two isn’t just because ” ‘q’ is almost always followed by ‘o’. ” It’s because [qo] is always ligated, no matter which scribe is writing.
Karl,
Thank you so much for taking the time and trouble – I’ve taken note.
Also – I’m fortunate to be one of the safest places in the world at present. When democracy works, it works well. No fuss, no pother about masks or about vaccines. One of the lowest per capita mortality rates, I understand. No sky news is good news imo.
D.
Lisa: The idea that “qo” represents a single character, a digraph, or a prefix was discussed multiple times before. There are for instance some blog posts by Emma May Smith, or by Nick Pelling about this topic. The main argument against such an idea is that the removal of “q” will almost always result in a valid word. With other words “q” represents an optional element.
For instance beside “qokeey” (308 times) also “okeey” (177) exists,
beside “qokeedy” (305) there is “okeedy” (105),
beside “qokain” (279) there is “okain” (144),
beside “qokedy” (272) exists “okedy” (118),
beside “qokaiin” (262) there is “okaiin” (212),
beside “qokal” (191) exists “okal” (138),
beside “qokar” (152) exists “okar” (129),
beside “qokol” (104) exists “okol” (82) etc.
(see also Table 3: Usage of common prefixes , , , and in Timm & Schinner 2020, p. 10)
Strict positional constraints are typical for Voynichese (see Currier 1976). The most strict constraints did indeed apply to “q”. “q” occurs in 99.4 % of the cases at the beginning of a word, and the glyph after “q” is in 97.5 % an “o”. “q” also mostly occurs after a word ending in “y”, and in 3116 (59 %) out of 5290 instances “qo” is followed by “k”.
But the positional constraints for “q” are not an exception. For instance the glyph “n” is almost always the last of a word (98.7 %), and the glyph after a sequence of “i”s is almost always a “n” (97.4 %), “m” is almost always the last glyph of a word (95.1 %) and even of a line (70 %). The glyph before a sequence of “i”s is most likely an “a” (94.2 %). …
On first view ideas like “qo” and “iin” might represent single characters look reasonable. But such ideas wouldn’t help to explain all the other positional constraints.
Nick – or Rene –
I’m not clear about what is meant by the term “fake stars” for f.11v. Does it mean they are not associated with a new ‘line/paragraph’ as the others are? Or perhaps that they appear to Rene to be details added later, by some other hand? Or perhaps that their form is not the same as is given most stars on that page and/or other pages? Or something else altogether?
Ok – it looks as if no-one knows what ‘fake stars’ means, so I’ll ignore that and return to describing the odd-looking ones as ‘hastily written’.
Also, Nick – about your suggestion that the original septenion should be re-imagined as formed of two quires… I looked into the question briefly and what I found was that the septenion is, in fact, one of the strongest arguments in favour of your (now abandoned) Italian theory, especially if the Voynich hand is, as you’ve supposed, a humanist hand.
Breaking that quire into two – conceptually or actually – would only serve to erase the evidence – which appears to be a marker of Byzantine influence, and to locate the manuscript’s production in Italy or Spain or ‘somewhere southern’ to which those who fled to the south from Byzantine territories had come by the early fifteenth century.
As I say, I didn’t spend a great deal of time on the question and it may be that septenions are more common than appears so far. The only way to know certainly, either way, would mean a massive study because some major collections don’t describe the quires – as e.g. the British library doesn’t. So to do an inventory of septenions would mean weeks – no, perhaps years, of page counting. All I’ve done is find two examples of mss on membrane, in Italy. One is by a ‘humanist’ and the other Hebrew. Quite a few more septenions from Italy, Spain and Constantinople. It looks as if that’s the ‘septenion line’ for the western Med.
Not ‘too many bifolios’ – just the right number for a work made in Italy, Spain or ‘somewhere southern’ in the fifteenth century.
Cheers.
Metadata is always useful.
Wilfrid Voynich was a chemist and pharmacist.
He would have no problem finding the formulas for 15th century
paints. So that he could have faked an illumated manuscript.
It looks like it was colored with a cotton swab.
An art teacher may be able to tell if the line drawings
and text were made by the same person. What books from the
same lot did Wilfrid Voynich sell to the British Museum ?
Where did the other books go ?
Did Athanasius Kircher collect erotica ?
Xplor – faking a medieval painting is not as easy as you might think. If you were to try faking one, the first thing you might do is hunt for paint recipes, but how would you know, if you lived before 1931, the molecular structure of azurite gained from Armenia, as distinct from Badakhshan, and how would you know which was being imported into thirteenth-century London or Paris? The scientific and historical disciplines (not to mention the art-historical matter) hadn’t been developed back then. The information just wouldn’t be there.
Besides, if it’s supposed to be a fake – a fake what? Nothing in it finds a true counterpart in any area of medieval Latin art, history of textual stemmata so far as anyone has been able to discover in 110 years.
So either it’s wholly original (which it cannot be because the drawings show evidence of derivation from pre-medieval sources), or its the lousiest fake you ever saw. Or…
D. O’D : Or…que compadre?…
Diane,
Yes, this is the question I always had about the Ms. If it’s a fake, a fake of what? The biggest elephant in the room being for both trained and untrained eyes, the highly queer “rosette” pages. I have formed some general ideas over time about this, and riffing on Nick’s idea that the center sphere is in fact St. Marks, I would say my ideas are highly terrestrial; these things are not planets or “dimensions”. Perhaps working together we can figure these things out, whether real or fake.
Matt
Matthew – after reading a post by Nick in 2010 – where he called that drawing a ‘hurricane of oddness’ and let me know that apart from a few vague speculations and one acute observation, nothing had been done to understand it, I worked on it for a few months and was able to confirm speculations about its being a map, explain the orientation, and various telling details and so on, and then (via a mass of historical, archaeological and other information) that it wasn’t created in one go, but had three identifiable recensions the last evidently in the early fourteenth century. It is clearly not of Latin European origin and is neither east facing nor interested in Jerusalem or Rome. The makers apparently had no interest in western Europe, or didn’t need to include it.
This information being unwelcome – didn’t suit the current theories – the information was initially ignored for 18 months or so, then a flurry of efforts to create ‘alternatives’ which were more Eurocentric, but those obviously being inadequate, the powers that be decided to revert to the old ideas such as that there’s a ‘clock’ in it and so on. I thought that a pity because the range and nature of the Voynich map is a key to much of its content and locked in very well with information offered by the astronomical sections and plant-pictures.
No point is trying to force people to absorb unwelcome information, though, is there?
Mathew Lewis: Charles Dawson’s Piltdown Man was inarguably the ‘worst fake you ever saw’ yet it fooled many of the so called paleo. experts for forty years and he made not a farthing from it, that never being his intention. Why do it, would be the question on everone’s lips when it was exposed in 1953, by which time happy old Chuck was long dead and home free, so quite a joke on his detractors from the Society you ask me. Dawson was an enthusiastic amateur at the game and his many real talents had never been recognised by the establishment, thus with a little help from his mates decided to get pay back in his own devious terms. So Couldn’t our VM have been a variation of the same general theme, put into effect by someone like Mary Boole who had been treated similarly, thereby putting down her own ‘don’t get angry get even stunt. She had five clever daughters for artists and scribes, plus the talents of her son-in-law Wilfrid the biblio. to sell the con, with the provisor that he might make profit from it if so motivated. It’s still making it’s intended mark on the intended targets as can be seen a good century and more later.
Diane,
It isn’t as clear, to me as it might be to you what would constitute “unwelcome” information and I guess I might think about it for a while, talk it over with friends, etc. before presenting it. No force would be involved, hopefully, ever. My ideas are vague at best, as it is, though, yes that it might be a map centered around Venice, author uncertain(with a quite out of vogue potential candidate), though it could be something constructed outside Europe, certainly. The Silk Road trade, is definitely an intriguing historical concept. Just thinking out loud here.
Cheers,
Matt
Lapis lazuli would be known and sold by chemists. What they may
not know is when mixed with some binders (egg) in paint it turn to a
gray green.
Other Non Sequiturs
Richard Garnett, British Museum and Voynich Astrology
John Matthews Manly , University of Chicago letters from Wilfrid Voynich
Orlando M. Pilchard sardines in the ms 408
John Sanders,
For whatever reason my first reply to you disappeared into the ether. I have to agree in all candor that indeed the VMs may be a modern document. That it is an outright hoax still has to be determined. Especially considering as you yourself mentioned the people of a slightly “revolutionary” disposition like Mary Boole who surround it. It certainly could be one however.
Matt
Matt: I’ve found from experience, words like fake & hoax are more likely to trigger Nick’s ‘robot’ register than ‘Imposter’ which has a more cultivated feel to it and is just as meaningful for the same job …As for Mary Everest Boole, ‘revolutionary’ might be a little too severe in all fairness but, being an anti establishmentist of some standing could not be debated. Seems that after being widowed with five daughters to raise, she got up and took on work as a school librarian and gave unorthodox Boole algebraic lessons after hours to supplement a low salary which did not impress the all male teaching staff. On being put off for her audatious behaviour her only friend was Dickie who ran the British Museum library and who was to become her lifetime supporter and possible abetor in her pay back moves against the despised establishment. Despite set backs Mary was still able to pass on her unique tutoring methods to her own daughters and an extended family of translators, pharmacollogists, sketch artists, botanists and homeopaths which included in-laws like the Hintons, Stotts,Taylors and of course her new Polish? mentored student Wilfrid Wojznic…Must get myself a copy of ‘Hintons & Booles’.
I’m wondering whether our esteemed Nahuatl fraternity, the likes of Taylor, Tucker, Pannozi, Springer and the Comegys boys are at all familiar with Ethel Voynich’s favourite nephew George Hinton, known to his south of the border amigos as Jorge and to his colleagues in the service as Mr. H. Back in the first half of the last century whilst working as a mining engineer, JH spent some of his free time collecting native plants, many of which he sent to Kew Gardens in Chelsea via Aunty Ethel. Along with son James he rode on horseback throughout the arid Mexican desert country and by so doing managed to name somewhere in the order of 700 native plant forms not previously catagorised. His short form wiki biography suggests most of the identifications comprised species collected in the twenties, thirties and early war years, though he had been active as a youngster going back to the turn of the 19th century, this according to Harvard faculty of Botany records under the Hinton name. Cut a very long story short (there’s so much more), by the time of his departure in 1947, he was rightly considered the foremost world authority on Meso-American native plant life…It don’t mean for a New York minute that VM can be traced directly back to George Hinton in any direct form…unless we like to include certain indirect links involving perhaps John Manley, Edith Rickart, Bill Friedman and the gang from MI 8 in The Big Apple if we want to go all the way.
John,
I guess I think if it was a modern hoax, that someone would have gotten wind of it by now and that hasnt happened. Plus the fact of just how easy is it to obtain “old” unused parchment? If someone wants to prove such a thing, they are going to have to have to source that material. Anyone out there have any unused vellum sitting around? I would really like it if Yale or someone would say, “no we are certain this is an antique document”.
Matt
Matt: We’re not going to make any progress rehashing tiresome junk like ‘unused old velum’ or worse still ‘that someone would have got wind of it by now’. Your re gurtitating that sort of crud is not useful and a little insulting, somewhat akin to raising issues about the SantaLacoma lensed telescopic medicine boxes and pangolin scaled armadillos. Time to get with the real issues like the many doubts still unsettled on provinence and the conveniently overlooked problems about technology centuries in advance of early fifteenth century knowhow &c., etc.
John Sanders,
Insulting? I honestly had no intent to do so in this case, though I have made my discontent known in the past on things. No, honestly I have personally sensed a pendulum shift regarding in what the *consensus* opinion is about the manuscript … to being a modern document (probably hoax), which is why I wrote you to begin with. Also, this is what you suggested, mentioning Mary Boole. I thought.
I think we are in for the long haul on this, and if anything my message to you was a plea to *everyone* to not be trendy. Think for yourselves.
Enjoy though. If you like mysteries, I still think this is one of the biggest.
Matt: Sorry, patronising might have been a better choice of words, for I’m never offended by all the tired old cliches thrown my way about Greg Hodgins’ velum dating, based on five non random slivers taken from VM and little accompaning detail eg., isotopic dietry transfer of an unknown foodchain etc. And as for the old ‘someone would have gotten wind of it by now’ attempted put off, not necessarily, not if the scheme was conceived as a family entwrprise with each of six members taking an active part in the input and compilation. You’ll recall Currier advising us that at least two, even up to seven or even eight scribes could have been involved. Anyhow my own position is firm and I’m not a fellow who makes a mystery any more mysterious than it really deserves, case in point the VM imposter of around 1903. Cheers mate js
John,
I must say, although I have followed your comments with sincere interest, you caught me off guard with that one. Its honestly about 2 or 3 bridges too far for me to believe, since I am well, just little old me, an amateur historian at best….that the dating is really that off. I think honesty you made a few people really happy by declaring this, and maybe some friends, Im guessing people have considered “what if” scenarios, off by a hundred or so years either way, (I know I have) though I have seen from reading this lists VM comments rather faithfully, there are some who definitely dont share this opinion and think I might be too open minded going the other way. It could possibly progress along the lines you indicate. Who knows, perhaps it will. I think quite a bit of explaining would be done.
Matt
Matt – the problem with many Voynich theories, and theorists, is that they never weigh the evidence for and against an idea before adopting it.
If I were to win a lottery, I’d send a subscription and the last twenty years’ back copies of the artefact conservation journals to all the (few) believers in the ‘not-genuinely fifteenth century’ theory.
They could learn about materials science, and the fine studies done to differentiate paintings on the basis of exactly which mines produced the raw material ground to make a particular pigment in paintings of a given date in a given locality. And they could also learn about XRF and all the other modern means employed today.
In the meantime, I just have to hope that one of them thinks to look as skeptically at what they believe, as at what they won’t believe.
Diane: A breath of fresh air if I may begrudgingly say so; allbeit in most fervent though respectful disagreement. Par for the course of course from a dyed in the wool and totally biased, fifteenth centuary timeline denier.
Seems that quite a few of the old Voynichero 1404/38 date line fellow travellers, may have struck their tents and drifted off to safer camping grounds, if that’s not too unlike their stength in numbers style. Talking of course about Rene, JKP, the two Thomas’ and a host of other past tense regulars. Surely they couldn’t all have gone to ground. Is it possible they’ve detected bad vibes eminating by way of a drastic change in perspective via Benieke library or Heaven forbid, Greg Hodgins and his somewhat shaky pigment back up out of Chicago. Just a thought and certainly no need for alarm…leastways not just yet.
Sanders,
You can add Jacques Guy to that list too. Whatever became of him? He had sooo much to say on the old list back in the day.
Matt
I thought I should mention I had a great deal of respect, and still do for Jacques Guy. Maybe he’s still out there? Pretty much everyone of substance who has stuck their neck out as well, and presented their evidence. even if they are forceful …and we are a “forceful” group of individuals here, that I do indeed respect. After photoflos and multiple better scans of the document, what do we have as far as translation? Not nothing, but not much either. A poll would be interesting to conduct as was done in the past to see where people are at in regard to what they believe it is, now. A little daring, as many of you are very guarded indeed I believe in tipping your hands. Makes sense to me for the most part.
Matt