This article was my April Fools joke for 2015, kept here for personal history and amusement. No, I don’t believe that René Zandbergen forged the Voynich Manuscript. And don’t worry, I’m still sane!
Introduction
In 2008 I became interested in the Voynich Manuscript and have been reading about it ever since. Over time I read many theories and went through different theories of my own, but one thing always got in the way – it just didn’t make sense. Like the mythical hydra, solving one issue would just raise more. Basically: the more you know, the more you don’t know. How could we possibly explain this artifact where no theory covers everything and the facts can contradict? Every now and again I felt a nagging gut feeling that something just didn’t add up. Something is just fundamentally wrong about the situation that I can’t put my finger on. It’s something that everyone here is thinking but nobody wants to say. We aren’t just on the wrong track, we are in completely the wrong field.
Eventually I gave in to these intuitions and started afresh with a blank slate. I cleared away all speculation, binned my previous work, disregarded the big names and ignored any assumptions and preconceptions that were holding me back. It was time for the bare facts, and the facts only. I built these basic truths into a new explanation without trying to prove any theory, trying to gratify myself, or considering what theories were popular. The result surprised and disappointed me. But the truth is the truth, it just is what it is, and it doesn’t change to comfort anyone. If you can’t face opposing viewpoints and prefer to hide within your own comfortable theories, I warn you not to read further.
I tried so hard and got so far, but in the end it doesn’t even matter. In short: The Voynich Manuscript is a modern forgery by René Zandbergen.
Recently on the Voynich Manuscript mailing list there has been a kerfuffle over a supposed Athanasius Kircher booklet find in Minerva Auctions’ catalogue. You can find the details summarised here in Ellie Velinska’s blog (don’t worry, the April Fools’ joke at the bottom is on her part, not Minerva’s). I’ll continue with what I’ve dug up.
They were founded in 2006 as Bloomsbury Auctions Italia, a branch of Bloomsbury Auctions. They were acquired by Dreweatts in 2011, but as part of an agreement they remained Bloomsbury Auctions Italia until 2012 when they were rebranded as Minerva Auctions. Their location and staff have remained constant throughout.
I’ll just briefly summarise some of the following discussion here to keep you up to date. Hypotheses fall into these categories:
Digital forgery. No item exists.
Item exists as modern forgery.
Item exists as historical forgery.
Item exists as historically genuine booklet.
Forgery could be a scam, PR stunt or prank, and it could be on the part of the seller or Minerva Auctions. Some details of the image are suspicious (diagram made by Rich SantaColoma who was the first to suspect a digital forgery with evidence) but those points could have been attributed to bad digital resizing and restoration of a genuine image (David Jackson). Some say the Latin is incorrect (“Cronolandensem” for example), the printing is sloppy and the year 1669 is far too late as the printer Hermann Scheus was no longer active for Kircher (or even in general, note how his last known work was in 1653) and Marci was already dead. Also as they admit, the booklet does not feature in any Kircher bibliographies, libraries or museums. The previous owners mentioned could not be found in ancestry records. The description sounds suspicious with enticing mentions that Kircher was onto something, and possibly PR related to a book.
So far there are only two mentions of the booklet (in the auction itself and an article from 2nd June 2013), and only one image. No other publicity or marketing could be found.
Update: That article now displays an “updated at” date of 11th January 2015, just after our discussion. Odd.
(side note: I wonder what would happen if someone tried to bid for something at the auction if it really only existed as a digital image in their catalogue? Sounds like an excellent way to get convicted of fraud and/or false advertising, and I’m sure they know it)
The other Kircher booklet
Out of five Kircher items at the auction, three books sold and two booklets did not. I will mention the other unsold booklet, Arcanorum Theatro, because it has interesting similarities to the main booklet in question.
Frontispiece of Arcanorum Theatro. Click on image for full size.
Translation of description (using Google Translate, sorry, I don’t know Italian):
339. Kircher, Athanasius – OfLodovico Locatellionce againimpressed onsome observationsof mysteriousTheatroin Venicerecentlyproposedby Athanasius KircherRome, from theprintingof LouisGrignani, 1670.In8th.30pp. Friezechalcographicthe frontispiece, title page in red and black, threeincisionsin the text, light gorathe firstpaper and smallwormholesto theback plateof the binding,missingthefinalwhite, Full parchmentcoeval, withthe backpiece to thetitlein gold.BookplateschalcographicPaulineElsonBright (Boston).
€3500 –€ 4500
Lotnot sold
The firstedition of the HandbookofalchemicalLocatelliwasengravedin 1644 byGiovanniPietroRamellati. Inthis raredissertationKircher, not present inanybibliographyconsulted andabsentfrom the richfundsofalchemical-hermetic “Joost R.RitmanLibrary” in Amsterdam, the British LibraryandtheBodleianLibrary, the Jesuit takesbutstrangelyexaminedonly the secondedition of the HandbookofLocatelli, printed in Venicein 1667by PaoloBaglioni.
It’s also very short (only 30 pages), could not be found in any libraries, and the previous owner doesn’t seem to appear in ancestry records. The similarity in style and image is uncanny. It even has exactly the same flaws/artifacts in the full digital image (here’s the link to Rich’s analysis of the original booklet again), and more.
Photoshop guidelines previously embedded in the image file, showing up as soon as I opened it for the first time.
When I opened the original image in Photoshop, to my surprise it already had guideline metadata embedded in the file. You can see them above; they align with the top and bottom of two of the text lines. (actually I was more surprised that JPG files could store that, but that’s beyond the point 😉 )
Text suddenly going blurry in Arcanorum Theatro.
The text suddenly goes blurry at the end of the word “Arcanorum”, why?
Upper left corner of Arcanorum Theatro with cut-off area indicated. Click on image for full size.
Zooming into the upper left corner, there is a weird rectangular area where the dots just cut off.
How the clone stamp was used in the corner of Arcanorum Theatro.
Viewing it at medium size makes the situation clear. The cut-off boundary aligns with the very upper-left corner. The shape of the stain was simply duplicated, and both copies overlap with each other. Appears to be a bad application of Photoshop’s clone stamp tool.
Identical bush emblem in Ars Magna Sciendi (left) and Arcanorum Theatro (right).
Lastly, the bush emblem is perfectly identical to that in Ars Magna Sciendi, which is known to be real, and possibly La Chine Illustrée, if that indicates anything.
Well if it’s a scam or PR stunt, it has a decidedly unsexy description. If it’s a joke, it’s not very funny. Whatever’s going on, I think it’s related to the main booklet in question. I talked to Ellie about this and we compared the two booklets.
Both Kircher booklets compared
With all of that in mind, looking at both unsold Kircher booklets together reveals a very strong connection.
Stains in the Kircher booklet with shapes outlined in red. Click on image for full size.
It turns out that all of the stains in the booklets are the same shape, just resized and rotated. You may point out that the one on the left in Arconarum Theatro doesn’t fit, but looking closer, we can see why.
Close up of stain in left of Arcanorum Theatro with shape (red) and cut-off boundaries (green) indicated.
There are more boundaries where the dots cut off too cleanly. When they are traced out, we can see that the “stain” was originally in the same shape as the others, but parts were erased.
Comparison of the footers of the Kircher booklets. Click on image for full size.
Looking at the footers together, we can see that the horizontal lines (including the supposed printing flaws) are identical, only resized a bit. “ROMAE” and “SVPERIORVM PERMISSV” also look suspiciously similar.
Comparison of the backgrounds of the Kircher booklets, with one flipped upside down. Click on image for full size.
Last but not least, putting the booklets side-by-side makes it abundantly clear that the “paper” of both are exactly the same background image, but one was turned upside-down! Now you cannot unsee it! The only difference is that it has a light blue tinge on the right in Arcanorum Theatro.
Conclusion
From my investigation and discussions, I think I can safely conclude that both booklets have always been modern digital forgeries. Yet I have also found that Minerva Auctions has a clean record. So who is behind it?
The most likely scenario, according to David Jackson and I, is that the forgers produced physical items from their digital files and attempted to sell it through Minerva Auctions. The forgers submitted the original files (not knowing the JPG would retain the Photoshop guidelines) instead of taking photos of the items. Minerva Auctions got suspicious after consulting the experts and libraries, figured out the fraud and blacklisted the seller. However by then the items were already printed in the catalogues so they couldn’t just purge it from their records. So now they just say “this item was already sold, stop asking about it anymore please” to cover that they accepted and nearly sold forged items.
However, the scenario does suppose a bizarre (and perhaps implausible) lack of competence on the part of both the forgers and the evaluators of Minerva Auctions. Also if Minerva Auctions discovered the fraud, it would have been far easier to just delete it from their website and say “sorry, the seller changed their mind and withdrew this item” at the auction. It is just a hypothesis so take it with a grain of salt and please don’t bother anyone about it!
(of course, if Minerva Auctions was telling the truth that they sold the booklet about the Voynich Manuscript, that means they actually have sold forged items…)
UPDATE (April 2015): Ellie says “After being confronted by several experts Minerva Auctions admitted that they themselves created the two ‘possible’ Kircher editions. They explained that there is a long tradition of ghost editions in the auction catalog and the two entrees are in line with this tradition.”
Here is Bunny’s find of possible hidden numbers and letters in the tree on f71r of the Voynich Manuscript (link goes to original image). It is reproduced on my site with permission (by request, in fact).
Possible hidden letters and numbers in the tree in f71r.
Adjustments in the image: “changed contrast brightness, gamma, colour then made b/w. attempting to remove green from tree and clarify what left, no adjustments made to actual lines of image.”
I was in a statistical geography mood so I made this map based on Wikimedia statistics. It shows the most popular Wikipedia language edition for each countries and territories that had hits in 2014 Q1. If the majority of hits from a place are for a single language, I marked that language’s colour. If there was no majority language, I marked the top two in a gradient.
I hope you find this as interesting as I did.
Map of countries and territories by most popular Wikipedia edition (2014 Q1). Click on image for full size.
Interesting points:
Out of ~6000 languages in the world, only 32 (0.5%) account for most popular Wikipedia edition in every country and territory in the world that tried to access it. All of these languages are from Eurasia, which really says something about the power structures over history and the digital divide.
Language geography corresponds well with European imperial holdings with some exceptions. Who would have guessed that Puerto Rico, Suriname and East Timor would have English as their preferred Wikipedia language? Regionalisation is also a factor.
English has more popularity than the rest of the languages combined.
Regions with no single majority language include North Africa, the Caucasus, the Balkans and the Baltics. Other such places include Belgium (French and Dutch), Norway (English and Norwegian), Greenland (English and Danish), Israel (English and Hebrew) and South Korea (English and Korean).
Leave your thoughts in the comments section below!
Every now and again we uncover manuscripts with possible direct or indirect links to the Voynich Manuscript. They might contain a similar glyph, a similar illustration, or perhaps a similar diagram. A good example was Cod. Sang. 839 (discovered by Thomas Sauvaget) with the same quire number style.
Cod. Sang. 754 is perhaps special in how many similarities there are.
All credit to the discovery goes to Job (from the Voynichese project); I am simply documenting it for him. I will avoid making any bold claims and simply lay out all the similarities and let you make your own decision. I’ll also not bore you with the details of the manuscript until the end.
1. The illustration
The first thing Job noticed was the style of the illustration on page 164.
Page 164 of Cod. Sang 754.
It should speak for itself.
(it’s the only full plant illustration in the manuscript so don’t bother looking for others)
This paper proposes a new pattern in the text of the Voynich Manuscript named the “Curve-Line System” (CLS). This pattern is fundamentally based on shapes of individual glyphs but also informs the structure of words. The hypotheses of the system are statistically tested by two independent people to judge their significance. It is also compared to existing word structure paradigms. The results suggest that the shapes of glyphs affect their placement in a word, the Curve-Line System is an intentional feature of the text design, and the text of the Voynich Manuscript is a highly artificial system.
My page for public Voynich resources is now successfully up and running, and on the main link menu. I’ll be posting all sorts of useful things in the future. Watch this space!
According to my latest cipher theory, this is a general estimate of what the encipherment process could be:
(Optional) Prepare your plaintext by removing some letters. Helps to save time.
(Optional) Split into blocks of equal length. Helps to reduce errors.
Convert each letter into a number with simple substitution.
Do some mathemagics to the numbers with Pascal’s Triangle (exact details are trade secret). You now have Voynichese! If you split the plaintext into blocks, each one now corresponds to a line of ciphertext. But some may have ended up with wildly different lengths, so…
(Optional) Pad out lines with filler at the beginning or end to make them equal length. Helps to make the result look nicer and harder to decipher.
