The Voynich Ninja

The British Library just posted a blog post with a few examples of medieval ciphers:
You are not allowed to view links. Register or Login to view.

Thanks. Learning the mindset of medieval scribes might prove invaluable for reading the vms.

Nice find Koen, thanks for that. - I agree with @lelle and would add that the mindset is of course directly related to the state of development. One can only apply what is known. This is actually a trite statement, but some "impressive solutions" are very far from it.

The replacement of vowels by nearby consonants was widely used, and is known as "Notae Bonifatiii".

It goes back at least to the 10th Century, but I don't know enough about the topic to be sure whether they are really related to St. Boniface, as one can read in various publications.

(24-03-2018, 10:29 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.The replacement of vowels by nearby consonants was widely used, and is known as "Notae Bonifatiii".

It goes back at least to the 10th Century, but I don't know enough about the topic to be sure whether they are really related to St. Boniface, as one can read in various publications.

It is a tradition that goes back to Hrabanus Maurus, de inventione linguarum, Bonifatius is accredited with two sytems of cryptography, 1) replacing vocals by a system of dots and 2) replacing vocals a e i o u by b f k p x, as far as I can see this tradition is acvcepted as fact, because there are contemporary English examples, the replacement by consonants seems to go back to Antiquity. There are many examples from the 9th c., of both methods in the Althochdeutsche Glossen

With respect to other substitution ciphers, this method is interesting because it is "lossy": the text is actually degraded since, for instance, the symbol 'f' in the encrypted text can read 'f' or 'e'. This makes the cipher somehow similar to an abjad: information is lost, but someone knowing the original language has no difficulty reconstructing the original.

BTW, this cipher reduces both first order (character) and second order (digraph) entropy: the effect on conditional entropy (the difference between second order and first order entropy) is minimal.

You anticipated my entropy questions,  Marco   it's certainly an interesting cipher in this regard.

It also shows that a text with information loss can still be read today.

Snap, Koen.

I noticed that too, and sent off a brief note to Nick Pelling, who responded by reminding me that he has already considered that cipher - first in a very brief post of 2008: "HBPPZ CHRKSTMBS TP YPX BLL!"


     You are not allowed to view links. Register or Login to view.

What one finds with the majority of medieval ciphers is that they are fairly simple (and often very simple) substitution codes, regardless of whether they are substituting letters or numbers.

As with the one that only substitutes consonants for vowels, sometimes they are relatively transparent (can almost be read without converting the letters).


Even Hildegard von Bingen's cipher (the substitution part, not the glossary part) is relatively easy to figure out. If you look at the letters for a while, you notice many of them are derived from the shapes of the letters they represent, simply flipped or slightly modified in a way that is easy to memorize.

The reason being, that today we are far more fluent in language and adept at switching between them than our ancestors were.
When literacy was highly restrictive, when reading was reduced to pronouncing phonemes aloud in prescribed fashions, these linguistic games were truly impenetrable except to the occasional well financed diplomat.