Earlier this year, I proposed that the VMs might possibly have brought by Guillaume Pellicier from Venice to Guillaume Rondelet in Montpellier, and listed a short bibliography of titles that would be worth consulting to check this out. A few days ago, I found one of them on the Internet:

Correspondance politique de Guillaume Pellicier, ambassadeur de France à Venise 1540-1542, publiée sous les auspices de la Commission des archives diplomatiques par Alexandre Tausserat-Radel (1899).

Usefully, pp.700-704 of this contain an (admittedly fairly terse) inventory of Pellicier’s personal library of 332 books. However, only one of these sounds even remotely like a candidate for the VMs:-

Hospitalières de Montpellier, Fonds de l’Hôpital général, B. lU; copie.
323. Liber de astronomia., gr., script.,continens lib. XX

Unfortunately, Pellicier seems amply Greek-savvy not to have misdescribed the VMs as a “greek” text, so this seems rather like a dead end. Oh well!

pp.710-721 contains the text “Vita Guillelmi Pellicier, episcopi Monspeliensis, ad virum eruditissimum Bernardum Monfalconiuiu, Benedictinum“, which includes a little bit about Pellicier’s book-buying on p.713 (penultimate paragraph).

As an aside, it’s also possible that had Rondelet acquired the VMs by other means, a quite different fate might have befallen it. According to this 2006 aricle on Clusius (p.95):- 

Rondelet himself became very worried after the arrest of Pellicier, and according to Joubert, secretly burned those books in his own private library which he thought might land him in trouble with religious inquisitors.

All of which defensive biblioclasm brings to my mind the underground “Bibliotek” memorial in Berlin’s Bebelplatz, located exactly where the infamous Nazi book burnings began – this was nicely evoked near the end of Alan Cumming’s recent BBC Four documentary on the film Cabaret. The plaque reads: “Where books are burned in the end people will burn.”

A couple of upcoming Voynich Euro-novels for your brief attention: firstly, the Spanish “El caso Voynich” (i.e. “The Voynich Case”) from Argentinian writer and columnist Daniel Guebel. Though references to it seem to have been temporarily removed from Guebel’s website, it appears to be based on contemporary-ish Voynich research, and may even include some VMs images. Due out next month (November 2009).

And secondly, “Engels Fall” (i.e. Fallen AngelsAngel’s Fall” or “Angel’s Case”) by Helena Reich has a qualmless secret society leaving a trail of corpses across Prague, each with a tarot card signature… might a mysterious order of alchemists hold the missing pages of the VMs? Prague Post reporter Larissa Khek (the heroine of Reich’s earlier book “Watery Grave”)  is determined to find out [etc, etc]. Due out in two months’ time (December 2009).

Voynich researchers often laugh about “dain daiin”, a curious text pattern that often crops up in the Voynich Manuscript’s strange text. But I’ve just noticed an (arguably) even stranger pattern on lines 20 and 21 of page f42r:-

shol chol shoky okol sho chol shol chal
shol chol chol shol ctaiin shos odan

Here, it’s as though Voynichese itself is breaking down while trying to express some awkward low-level concept. If these two lines run on (as you’d expect), the mid-sequence runs “sho chol shol chal shol chol chol shol“. Sorry, but I really don’t buy into the idea that something as artificially structured as this could ever be some kind of repetitive pidgin or Hawaiian-style language (as those in the Voynich linguist camp would suggest): rather, these “words” more closelt resemble machine burblings, the output of some kind of proto-algorithmic process.

To my eyes, there’s a kind of elegant quasi-numerical symmetry to this, as if “sho[l] chol shol / chal / shol chol chol shol” is verbosely enciphering “X I X / or / X I I X“. Might these indeed be heavily enciphered Roman numerals?

I haven’t looked for this beyond f42r, but please leave a comment here if you find further examples elsewhere in the VMs! 🙂

Incidentally, the raw instance counts for chol /shol etc together with a rough percentage (showing how far against raw chance the combination occurs) are:-

  • chol – 780 –> 250%
  • chor – 501 –> 325%
  • shol – 278 –> 216%
  • shor – 152 –> 239%
  • char – 156 –> 87.5%
  • chal – 120 –> 70.6%
  • shar –   47 –> 64.1%
  • shal –   28 –> 40.1%

That is, if “ch” (5.66%) and “ol” (2.92%) sppeared randomly throughout the VMs, “chol” ought to appear roughly 311 times, whereas it actually appears 780 times – hence 780/311 = 250%. That is, the “chol / chor / shol / shor” set is 2.16x to 3.25x more likely than chance to appear, whereas the “chal / char / shal / shar” set are all less likely than chance to appear.

Having deconstructed the Beinecke’s own Voynich Manuscript page here the other day, I thought I ought to repeat the trick for my own Cipher Mysteries Voynich page, which was similarly in need of a bit of renovation (if not TLC). OK… so I actually rewrote it from scratch.

And now, here it is – Ladies & Gentlemen, for your delectation and delight I bring you… the most genuinely useful webpage on the Voynich Manuscript.

As normal, let me know what I’ve got wrong & I’ll fix it. 😉

German fans of Pater Castell and of the Voynich Manuscript have a treat in store coming up, with the episode due to be aired on 5th November 2009 at 20:15 called Das Voynich Manuskript (hopefully you can translate that from the German).

Here’s my rough translation of the programme blurb:-

16-year-old Ralf Hoffmann is found dead at Athanasius Kircher boarding school. Is it suicide? Father Castell, a former student himself, is asked in by his longtime friend Father Jonas Selby because a tragic event from the past seems to have exactly repeated itself. Marie Blank is surprised when she sees a youthful portrait of Father Castell on display and is even more surprised by what she finds at the scene. The two quickly decide to investigate the mystery together. Ralf Hoffmann was a misfit interested in encryption and anagrams – and especially in the undeciphered Voynich Manuscript. Crucially, Father Castells’ knowledge helps Marie in her investigation. It turns out that others in the in boarding school share Ralf’s interest in anagrams and encryption. A secret school club is active, even blackmailing a teacher – might Ralf’s death be linked to this “Voynich Club”?

In 1931, John Matthews Manly (who was very sharp, both historically and cryptologically) pointed out that the Voynich quire numbers were written in a 15th century hand – you can tell this from the characteristic ‘4’s, ‘5’s, and ‘7’s. To be precise, even though a fair few of the VMs quire numbers appear to have been added later (most obviously the numbers for Q19 and Q20, but also those for Q6 and Q7) for reasons as yet unknown, the bulk of them are indeed 15th century.

unusual-quire-numbers
The four quire hands, from The Curse of the Voynich (2006) p.17

What isn’t widely known is that there is also a (quite different) 15th century hand in the marginalia. Even though some people like to dismiss the hard-to-understand writing on the back page of the VMs (f116v) as merely “pen trials” or “doodling”, I think you can look past the codicologically tangled mess to see that the earliest (faint) hand has distinctively 15th century letter-shapes, as underlined in green here:-

ahia-maria-annotated

Whatever all the other words on f116v happen to read (and people will no doubt continue to debate that for a fair while yet), I’m pretty sure that this faint (but apparently unemended) section reads “a + ma+ria“, and that it is (from the distinctive shape of the three ‘a’ characters) written in a 15th century hand, one quite different from the quire numbers.

What does this tell us? Given that the Voynich Manuscript almost certainly turned up in Rudolf II’s Imperial Court in the first decade of the 17th century, and assiduous archival trawling has turned up no definite earlier reference to it, I believe that this points to two main scenarios to choose from:-

  1. [Real] It’s a genuine mid-15th century object.
  2. [Hoax] It’s a (probably late-)16th century fake, designed to resemble a genuine mid-15th century object.

To which I would further add that Voynichese is apparently designed to look like an enciphered 14th century herbal (i.e. written in a medieval simple substitution cipher, with medieval herbal illustrations, and medieval page references), even though the parallel hatching and handwriting are both 15th century in style. All of which suggests three scenarios to consider:

  1. [Real] It’s a genuine mid-15th century object designed to resemble an enciphered 14th century herbal.
  2. [Clever Hoax] It’s a (probably late-)16th century fake, designed to resemble a genuine 15th century object designed to resemble an enciphered 14th century herbal.
  3. [Dumb Hoax] It’s a (probably late-)16th century fake, designed to resemble an enciphered 14th century herbal, but with a number of 15th century details included by mistake.

All the same, is it really the case that one individual late-16th century hoaxer / faker was sophisticated enough to add multiple 15th century hands to the quire numbers and back page? Well… possibly: but it should be no surprise that I think the historical odds are very much against it. Your mileage may vary, of course.

(As an aside, it has recently been suggested that the VMs might have come from around 1300 (and I shall soon be posting about Patrick Lockerby’s series of VMs-related posts): but the presence of parallel hatching in the VMs would seem to be a strong indication that even 1400 would be too early a date.)

A fascinating email just arrived at Cipher Mansions from Tony Gaffney, our virtual cryptologer-in-residence at the British Library. While looking at BL Add. MS 39660 recently, he noticed a set of dates for ten popes written in an unusual mixture of Roman numbers and Arabic numerals (“an9 pm9” = “annus primus“, and “ufq3” = “usque“):-

That is:-

  1. cclxxxij
  2. m cclxxxiiij
  3. m cc lxxxx
  4. m cx ij
  5. 1 40 viij
  6. 1 4 10 an9 pm9
  7. 14 12
  8. 14 17 ufq3 1430  an9 pm9
  9. 1 431 ufq3 1446
  10. 14 46 ufq3 1455

According to the BL’s bibliographic description, this was written on paper in Italy, with the later popes added not before 1455: while Tony adds that the “v” in the fifth date “is written in the old style of a backward sloping b“, hence a 15th century hand. All of which gives us a basic prediction for where and when we might expect to find this unusual kind of mixed Roman / Arabic numbers: Italy in the second half of the 15th century. Examining BL Add. MS 39660 even more closely may to help us be more specific: but that’s a job for another day.

The presence of “pm9” here is particularly heartening, as this is precisely what is used for the quire number in Q1 of the Voynich Manuscript. Intriguingly, Tony notes that the “cc” pairs in the first three dates are ligatured at the top, just like the EVA “ch” glyph, though he has previously seen this in the 14th century Royal MS 12BXXV f.283 (which is a “table converting Arabic & Latin numbers“). And furthermore, he adds that “in the fifth [date], we have not only a combination of Arabic and Latin numbers but the 4o is the Voynich EVA qo linked!!“. Of course, that might just be a coincidence, but even so…

Given that the Voynich Manuscript is owned by the Beinecke Rare Book & Manuscript Library, you’d perhaps expect its online description of the VMs to be sober, accurate and helpful – a useful antidote to the speculation-filled Wikipedia VMs page.

Unfortunately, it isn’t.

As a technical writing exercise, I thought I’d dismantle its description to give a more accurate picture of where sensible Voynich research now is…

Written in Central Europe

Hmmm… because the pictures (Italian architecture) and the zodiac marginalia (Occitan) both seem to point to Southern Europe and I can’t really think of any evidence that specifically points to Central Europe, this is hardly an encouraging start to the whole page. Oh well…

at the end of the 15th or during the 16th century,

Given that John Matthews Manly pointed out 75 years ago that the VMs’ quire numbers were written in a 15th century hand, and that we are now quite sure that these were not original, I think “or possibly during the 16th century” might be more balanced (basically, to throw a sop to the vocal hoax and Askham clans).

the origin, language, and date of the Voynich Manuscript—named after the Polish-American antiquarian bookseller, Wilfrid M. Voynich, who acquired it in 1912—

Polish-Anglo-American” would be more accurate, as would “who claimed to have acquired it in 1912” (Voynich was never completely open about how he bought it).

are still being debated as vigorously as its puzzling drawings and undeciphered text.

Fair enough. 🙂

Described as a magical or scientific text,

…as well as a heretical, alien, channelled, medical, or nonsensical text (unfortunately). Not really a helpful clause, so probably should be dropped.

nearly every page contains botanical, figurative, and scientific drawings of a provincial but lively character,

As the paragraph then goes on to categorize the drawings, reducing this to “…contains drawings of a provincial but lively character” would probably be an improvement.

drawn in ink with vibrant washes in various shades of green, brown, yellow, blue, and red.

This isn’t particularly accurate: while some colours are indeed vibrant (redolent of 16th century inorganic paints), some are actually very faded (redolent of faded organic washes). Describing them all as “washes” also misses out the entire “light painter / heavy painter” debate that has been ongoing for some years.

Based on the subject matter of the drawings,

Rather too simplistic: “based on the apparent subject matter” would be more correct.

the contents of the manuscript falls into six sections:

Again, this doesn’t really do justice to the nuanced view that Voynich researchers now take: which is that the names of the sections are mainly useful as a means for referencing them, whatever the actual contents ultimately turn out to be. Hence, I would replace this with “Voynich researchers group the pages of the manuscript together into six categories”.

1) botanicals containing drawings of 113 unidentified plant species;

Actually, Voynich researchers prefer to call these “herbal” pages, because European botany (in its modern sense) only really began in the 16th century with Leonhart Fuchs and (arguably) Ulisse Aldrovandi, hence the term “botanical” might well be anachronistic. Furthermore, “unidentified” isn’t really true, since there are a handful of plants (most notably the water lily on f2v!) about which nobody seems to argue. So, “1) herbal pages containing drawings of 113 plant species, most of which are unidentified” should be preferred. Also, this omits from the count the second set of herbal pages in Q15 and Q17: and even adding those would fail to notice that some of the herbal drawings are apparently duplicated on different pages (most notably f17v and f96v, but there are others). So, “113” is a bit of a questionable number: I’d prefer “more than 120″.

2) astronomical and astrological drawings including astral charts with radiating circles, suns and moons, Zodiac symbols such as fish (Pisces), a bull (Taurus), and an archer (Sagittarius), nude females emerging from pipes or chimneys, and courtly figures;

Again, Voynich researchers not only prefer to call these “cosmological” and “zodiacal” pages, but also normally split them up into seprate sections. “Astral charts” isn’t really certain, so perhaps “circular diagrams containing stars” would be more representative. The Sagittarius “archer” is actually a crossbowman, which (yet again) has a debate all of its own. A good number of the zodiac nymphs are clothed rather than nude (particularly in Pisces), only a minority are placed in “pipes or chimneys” (which might equally well be maiolica albarelli), and not all of them are female.

3) a biological section containing a myriad of drawings of miniature female nudes, most with swelled abdomens, immersed or wading in fluids and oddly interacting with interconnecting tubes and capsules;

These days, Voynich researchers generally prefer to call Quire 13 the “balneological” section (though I myself sometimes just call it the “water” section), because “biological” seems rather to be prejudging the contents. Again, I prefer to call the naked figures “water nymphs” rather than “nudes”, as this fits in with the general water / bathing theme, and also serves to separate them from the (quite different) zodiac nymphs.

4) an elaborate array of nine cosmological medallions, many drawn across several folded folios and depicting possible geographical forms;

We prefer “rosettes” to medallions; they are all drawn across a single 3×2 fold-out sexfolio, and would be more accurately described as “apparently depicting architectural and geographical forms“. Calling them “cosmological” seems unnecessarily presumptuous.

5) pharmaceutical drawings of over 100 different species of medicinal herbs and roots portrayed with jars or vessels in red, blue, or green, and

The term “pharmacological” has long been preferred for these: and there is an ongoing debate (hi, Rich) about the wide range of jars and vessels depicted.

6) continuous pages of text, possibly recipes, with star-like flowers marking each entry in the margins.

Personally, I’d say they’re more likely to be “flower-like comets” (i.e. some kind of pun on “caput”) than “star-like flowers”, but who knows? And they apparently mark the start of each paragraph (i.e. chapter / caput), rather than an “entry”.

History of the Collection

Like its contents, the history of ownership of the Voynich manuscript is contested and filled with some gaps. The codex belonged to Emperor Rudolph II of Germany (Holy Roman Emperor, 1576-1612), who purchased it for 600 gold ducats and believed that it was the work of Roger Bacon.

This doesn’t really summarize Marci’s letter to Kircher at all. Though Marci had heard these things, he didn’t know if they were true (and he seems keen to distance himself from the Roger Bacon claim).

It is very likely that Emperor Rudolph acquired the manuscript from the English astrologer John Dee (1527-1608). Dee apparently owned the manuscript along with a number of other Roger Bacon manuscripts.

No: although Wilfrid Voynich quickly took the view that this is what must have happened, it is actually very unlikely.

In addition, Dee stated that he had 630 ducats in October 1586, and his son noted that Dee, while in Bohemia, owned “a booke…containing nothing butt Hieroglyphicks, which booke his father bestowed much time upon: but I could not heare that hee could make it out.”

Even though this is a pretty slim pair of reeds to construct a castle upon, that hasn’t stopped plenty of would-be builders since Wilfrid Voynich trying.

Emperor Rudolph seems to have given the manuscript to Jacobus Horcicky de Tepenecz (d. 1622), an exchange based on the inscription visible only with ultraviolet light on folio 1r which reads: “Jacobi de Tepenecz.”

Actually, it reads rather closer to “Jacobj z Tepenec“, and there is also a deleted “Prag” beneath it.

Johannes Marcus Marci of Cronland presented the book to Athanasius Kircher (1601-1680) in 1666.

Once again, Marci tried to present the book to Kircher in 1665 (not 1666), but we have no evidence it actually arrived. Other cipher pages sent with correspondence to Kircher have disappeared, though: all in all, the manuscript’s precise provenance for the next century remains something of a mystery.

In 1912, Wilfred M. Voynich purchased the manuscript from the Jesuit College at Frascati near Rome. In 1969, the codex was given to the Beinecke Library by H. P. Kraus, who had purchased it from the estate of Ethel Voynich, Wilfrid Voynich’s widow.

Actually, Hans Kraus bought it from Anne M. Nill, who had inherited it from Ethel Voynich.

References

Goldstone, Lawrence and Nancy Goldstone. 2005. The Friar and the Cipher: Roger Bacon and the Unsolved Mystery of the Most Unusual Manuscript in the World. New York: Doubleday.

Romaine Newbold, William. 1928. The Cipher of Roger Bacon. Philadelphia, Pennsylvania: University of Pennsylvania Press.

Manly, John Mathews. 1921. “The Most Mysterious Manuscript in the World: Did Roger Bacon Write It and Has the Key Been Found?”, Harper’s Monthly Magazine 143, pp.186–197.

Really? A Voynich bibliography without Mary D’Imperio’s “The Voynich Manuscript – An Elegant Enigma”, without Jean-Claude Gawsewitch’s “Le Code Voynich” near-facsimile edition, and without (dare I say) “The Curse of the Voynich”? Not very impressive.

In summary, then, it’s an article which (despite mentioning a 2005 book) seems to reflect the inaccuracies and fallacies of Voynich research circa 1970. I’d happily rewrite it for them – but is the Beinecke actually interested? I wonder…

What can I say? If you want to be completely literal about it (like XKCD fans), it’s a brand new theory about Voynichese being scallop language (with the top two lines of f15v translating as “I think you should stop browsing forums and get back to work“). Otherwise, you might want to riff on how the final space insertion cipher stage is particularly clear here, and how annoying it is that the second line (with its four consecutive “or” verbose pairs) is absent from the Takahashi EVA transcription. Or to discuss how the first letter of the second line should be transcribed (it’s not at all obvious). Or even how best to cook scallops. You choose. 🙂

voynich-scallop