Could the Voynich Author Be a "Ramanujan-style" Anomaly?

[Jorge_Stolfi]

It seems that the author’s knowledge, particularly the knowledge he presented in the book, was quite modern for his time. For example, he must have had some understanding of herbalism and astrology.

But those two subjects were not at all "modern for his time".  Herbals and astrological handbooks/charts had been produced since Classical Antiquity, and were widely available by that time.

On the other hand, the plants in the Herbal section were not recognized by doctors at the tme (see Baresch's letter to Kircher), and the structure of the Zodiac diagrams seem quite original.

Perhaps he wanted to learn more about anatomy. However, back then, knowledge was rather limited, and so the author could explain it in his own way.

My impression is that the "anatomical" images in the Bio/Bathing section are more the product of the Author's guesses and/or fantasies than based on first-hand observation or copied from books of anatomy that existed at the time.

What about the cipher? Well, it’s most likely based on homophonic ciphers of that time, just made more complex in its own way.

AFAIK, by that time (before 1450 or so) there were hardly any books on cryptography.  Governments employed cryptographers, but the few ciphers I have seen described here were only complex enough to resist cracking with the methods known at the time.  

It is always possible that the VMS Author invented on his own a cipher so hard to crack that hundreds of modern cryptographers could not get even a tiny clue about its nature.

However, I don't think that private individuals like scholars and merchants would have felt the need to use anything much more complicated than a letter substitution cipher with homophones and nulls, plus some abbreviations.  Especially to encrypt a whole book, instead of a letter.  

In particular, I cannot believe that the VMS was encoded with a codebook cipher (where each word is replaced by an arbitrary string, like a number, according to a separate dictionary).  Decription of such a text is way too slow. 

As for schemes like the Naibbe cipher (that encodes each letter of the original with a whole Voynichese token), they would have the same problem, and additionally would waste too much time and vellum (at least 6 times as much as a simple letter substitution cipher).  Besides, the labels seem to exclude any verbose cipher scheme.

All the best, --stolfi