(20-03-2019, 07:41 PM)Helmut Winkler Wrote: You are not allowed to view links. Register or Login to view.@ David
You got the idea, but you have to begin to think like a medieval scholar, not like a modern mathematically minded scientist. The VMs writer did not construe a ciphre, but developed out of the medieval abbr. system, which was very flexible anyway in the 15th c., a kind of personal shorthand. And I think the problems you are adressing are pseudoproblems, no one, for example, knows anything about the statistics of an abbr. medieval Latin text, especially since the statistics must vary with every single ms.
Helmut, we may disagree with each other on a lot of details, but I agree with everything you say here 100%. In particular, ***think like a medieval scholar, not like a modern mathematically minded scientist***. Yes, exactly!!! Also, ***no one knows anything about the statistics of an abbr. medieval Latin text***. And further, ***the statistics must vary with every single ms.*** Yes, yes, and yes again. Exactly.
Now I happen to believe the ms is written in what I describe as late medieval Byzantine quasi-Judaeo-Greek. You believe it is simply some kind of combinations of medieval Latin abbreviation symbols. (By the way, it is not impossible that a scribe had the idea to write Byzantine Greek using medieval Latin abbreviation symbols.)
But the point is, in either case, the statistics will not necessarily look anything like any printed text. In the other thread I mentioned the example of Chaucer: just the famous word "soote" alone, the last word of the famous first line of one of the best-known works in English and world literature, is not among the *seven* most common ways to write that word *even in Middle English*! If we have to go to the 8th choice for a common word just to be able to read the first line of the Canterbury Tales, how much more so will we have to consider an extremely great variety of obscure variations of words, to be able to read substantial amounts of text of an obviously much more obscure work?
You know, there was an actual special form of printed type, developed in the 18th century, called "Record type". It was designed in order to publish a printed form of the Domesday Book that would represent the actual original script of the work as precisely as possible. I attach a couple examples of "record type" to this comment. Technically, accurate statistics on medieval texts would have to be done on this type of text!
(Rene will make the point that in these types of texts, the entropy values, etc., will be much, much higher, due to the great variety of abbreviations and combinations, and that they do not resemble the Voynich MS text, which has low entropy values because it has a very small and limited number of types of characters and combinations. This is true.)
In any case, the point is that a medieval ms by its very nature is very different from any type of printed work of any kind.