After spending too much time reading and collecting evidence on this book, I (tentatively) believe it is a medieval hoax.
Opinions on the artifact
Note that history’s not my specialty.
Where was it made? Alpine Europe at the confluence of Germany and Italy. Specifically, the Lake Constance area. Underlying content probably includes influences from other areas.
When was it made? Mid 15th century, circa 1460. Underlying content possibly created earlier.
Who made it? Some clever cookies.
Why was it made? Probably scam. Could also be esoteric or fantasy work.
Opinions on the text
Now for the real juicy bits I’m interested in.
Does the text have meaning? I highly doubt it.
What is the nature of the text?
15th century SCIgen analogue. Gibberish with an underlying generation mechanism.
If it has meaning: Transposition cipher, code, artificial language, non-textual nomenclature, or somewhere in between. I would bet that the trick lies in the creativity of the system rather than the complexity; after all there’s a limit to technology and practical effort.
I no longer believe it can be a natural language, nor any type of non-transposition cipher.
What does the text say? If it means anything, it’s probably about what Brévart calls the medieval “wonder drug” genre, with therapeutic uses of magical plants (with a great deal of medieval astrology and alchemy weaved in, no doubt). Sorry, no Atlantean alien time travel secrets here.
Now for something more personal,
What is my theory? Here is a brief history of my thoughts.
- 2008 to 2012 – Hoax: Concluded that it was a medieval hoax and just forgot about it, since there was no meaning to decipher.
- 2012 to mid 2014 – Natural language: After discussing with a friend, we agreed that it was in an unknown natural language. After investigating I was convinced that it was a Turkic language with some Arabic and Latin loanwords. I managed to translate a few words; for example the titles on f1r were “Natural crops”, “Classification of Western Zodiac”, “Fabrication system” and “Medicine”. I also managed to identify some plants this way, for example I thought the one on f6v was Liquidambar orientalis. I gave up on this idea when I found that many plants didn’t fit their supposed translation, and the text statistics rule out pure natural language.
- Mid 2014 to 2015 – Cipher: I figured out a possible word structure that suggested an encoding mechanism. Turns out this encoding would account for a lot of the observed statistics of the text. Neat, huh? I would best describe it as a clever code/syllabary combination. Applying it to the VMS I confidently placed its plaintext language specifically as Middle Low German and started decryption. However, further investigation revealed text properties (e.g. the CLS) that completely demolished my suggested system. Dang it.
- 2015 to present – René Zandbergen made it 😉
- 2015 to present (seriously) – Again concluded that it was a medieval hoax, except this time I have plenty more evidence (not ready for publishing yet though). I agree with David Jackson’s suggestion that the text generation was semi-mechanical and probably used combinatorial paper volvelles. My current idea is that there were multiple four-ringed volvelles with a shared glyph set. This system was created for efficiency’s sake, allowing multiple scribes to work in parallel, generating text that would come out looking consistent. This is hard to detect because the volvelles went in and out of use, the glyphs were in different positions and combinations on each volvelle, and many of the spaces on the rings were blank. Still, with the right methods, it should be possible to prove (or disprove) the theory and reconstruct the original volvelles. And there is still a human element to the text generation which I’ve yet to comment on…